Device-Related Infections: A Tale of Two Hospitals Setting Example for HAI Reduction, Prevention

We share the success stories of two hospitals that are driving down device-related infection rates and upholding high levels of patient safety within their institutions.

Sutter Roseville Medical Center
Sutter Roseville Medical Center (SRMC), in Roseville, Calif. is reporting that its vascular access team has gone nearly seven years without a central line-associated bloodstream infection (CLABSI) in any of its peripherally inserted central catheter (PICC) lines. The facility has recorded no CLABSIs, which are potentially deadly, in any PICC line placed by the team at Sutter Roseville Medical Center since January 2006, when a new bundle of central line practices and devices was implemented. A central line bundle is a collection of evidence-based practices and devices, and other best practices, designed to prevent bloodstream infections.

The accomplishment is notable for several reasons:
- This is the longest span without CLABSIs ever reported by a hospital.
- Central intravenous therapy (IV) lines, including PICCs, have a much higher risk for bloodstream infections than any other type of IV line.
- CLABSIs are frequently fatal, with an attributable mortality rate of 12 percent to 25 percent.
SRMCs record span without CLABSIs was reported in a scientific poster presented at the annual conference of the Institute for Healthcare Improvement (IHI).

Author Sophie A. Harnage, RN, BSN, VA-BC, who is clinical educator for infusion services, heads the hospitals vascular access team and supervised creation of the bundle.

The poster describes seven elements of the bundle that are key to preventing CLABSIs:
- Ultrasound-guided PICC placement, which aids proper location of the catheter.
- Maximum barrier precautions. Use of gloves, gown, mask, drapes, and related precautions helps prevent transmission of pathogens.
- Central line kit revision. SRMCs central line kit was expanded to include three new elements: a patient skin preparation solution containing chlorhexidine gluconate (CHG), a broad-spectrum antiseptic; a catheter stabilization device to help prevent catheter migration, insertion site trauma, and extraluminal migration of bacteria; and a foam disc that protects the catheter site and releases CHG for up to seven days.
- Change to zero displacement IV connector. Sutter switched the type of IV connector (a sophisticated plastic device that connects an IV catheter to tubing and helps deliver medications and nutrition). The hospital went from a so-called positive displacement connector to a zero displacement connector. Positive placement connectors are associated with higher rates of CLABSIs, according to several studies. The zero displacement connector, by contrast, prevents blood reflux (the backflow of blood into the connector). Blood reflux creates a breeding ground for dangerous bacterial colonization in the connectors interior pathway.
- IV connector septum disinfection. Nurses access connectors through an external part called a septum to draw blood or administer medications and other fluids. SRMC mandated thorough disinfection of the septum before every line access with a scrub the hub campaign.
- Standardized catheter flushing protocol. Sutter revised its flushing policy, requiring all central line catheters to be flushed with 10 mL of normal saline every eight hours and as needed.
- Daily monitoring. This daily site check of the line to evaluate its necessity makes it possible prevent complications and to remove central lines as soon as they are no longer needed.

I think the biggest piece of our success is the comprehensiveness of what we do, says Harnage. I cant point to a single element of our bundle that we could do without and still give our patients the protection they deserve. Nor can I imagine having this kind of success without a team who is passionate, dedicated, and committed to the delivery of safe patient care, from top to bottom. Everyone works together here to make our zero-infection goal an achievable reality.

The switch to a zero displacement IV connector is particularly notable because connector types are rarely addressed in central line bundles -- even though widely used connector designs such as positive and negative displacement connectors are associated with higher CLABSI rates.

The zero displacement connector used at SRMC was designed to avoid the pitfalls of previous connector types. For instance, the connector has a straight fluid pathway with no dead spaces so it is easy to flush clean. Other connector types can trap blood in the pathway, leading to bacteria colonization.

The poster also described several human factors that are crucial for preventing CLABSIs. These include catheter placement and care by a specialized vascular access team that owns the lines. Owning these lines, including initial assessment of the patient, insertion, care and maintenance, which includes daily site checks, has been the consistent practice of this Vascular Access Team, and a notable component of this central line bundle.

The poster presented at IHI reports that hospital-wide including all central lines, there has been one CLABSI since 2007, involving a dialysis catheter placed by a team from outside the hospital that did not follow the PICC teams protocol.
SRMC also promotes a patient safety culture that involves all interested personnel from top administrators to materials managers to infection control professionals and front-line clinical staff.

IHIs 24th National Forum on Quality Improvement in Healthcare was held Dec. 9-12 in Orlando, Fla.

Q&A with Sophie A. Harnage, RN, BSN, VA-BC
William R. Jarvis, an acknowledged world leader in hospital-acquired infection prevention, discusses Sutter Roseville Medical Center's accomplishment with Sophie A. Harnage, RN, BSN, VA-BC in the following Q&A:

Jarvis: Congratulations, Sophie, on a spectacular story of success in eliminating peripherally inserted central line (PICC)-associated bloodstream infections (BSIs). Can you tell us what proportion of your hospitals central lines are PICCs, for which you have primary responsibility -- versus other central lines, for which you do not have responsibility?  What percent of your patients are in the intensive care unit (ICU) vs non-ICU? 

Harnage: I should clarify that our vascular access team (VAT) has responsibility for all central lines at our facility, not just PICCs. We have comprehensive responsibility for PICCs, starting with assessment for the appropriate device and insertion, continuing through care and maintenance including declotting lines, weekly dressing changes, troubleshooting, educating, and line discontinuation.
We also led the way in developing our bundle. Its important to note the key bundle elements. They are:
 -- A maximum barrier precaution kit.
-- Ultrasound-guided PICC placement in the upper arm, with the basilic vein as the vein of choice. Use of ultrasound has become the standard of practice for all central line insertions.
-- A zero fluid displacement needleless connector (InVision-Plus®, RyMed Technologies) that replaced our previous positive-pressure IV connector. The literature associated positive-pressure connectors with higher rates of central line-associated bloodstream infections (CLABSIs). 
-- A central line dressing kit that includes CHG skin prep, BIOPATCH® Protective Disk with CHG (Ethicon) and Tegaderm (3M) dressing. Weve also added Statlock Stabilization Devices (Bard Access Systems), which eliminate the need for sutures.
-- Consistent manual disinfection of IV connector hubs, constantly reinforced through education and supervision. We call our education efforts the Scrub the Hub campaign, because the standard manual disinfection method is known as scrub the hub.
-- A standardized, saline-only catheter flushing protocol except for ports and dialysis catheters, which require saline and a heparin flush. We also check patency of all catheter lumens before usage as part of this protocol and re-establish patency when needed.
-- Daily monitoring of all central lines, using a data collection instrument that addresses the patient, unit, type of line, insertion site, and any identified complications.

Our team is responsible for assisting physicians with insertion of centrally inserted central catheters when requested. This collaboration not only prevents delays in therapy and treatment but ensures that all bundle components are being followed.
Regarding the percentage of our central lines that are PICCs, it averages about 90 percent to 95 percent of our total central lines, based on what I see on our daily communication board, where we track our central lines. Approximately one-third of the patients under our care are in the ICU.

Jarvis: Can you tell us what the CLABSI rate is for the non-PICC central lines in your hospital over this period?

Harnage: Your question seems to assume a difference between lines our team inserts and lines inserted by our physician colleagues. The VAT has responsibility for all central lines in one way or another, so the distinction between PICC lines and non-PICC central lines isnt meaningful at our facility. That said, the CLABSI rate over the past seven years is zero for all central lines either inserted or maintained by the VAT.  Our institution experienced one CLABSI in a dialysis line in 2007.  At that time, a dialysis agency with whom we had contracted independently managed all dialysis catheters.  After the CLABSI incidence, the agency adopted our bundle. There have been no further CLABSI episodes in dialysis catheters.  

Jarvis: How has your hospitals CLABSI rate in non-PICC catheters changed since involvement of your team and your approach?

Harnage: In 2004, our hospital had 13 CLABSIs housewide and in 2005, we had 11. In January 2006, we implemented our revised central line bundle and refocused the responsibilities of our VAT as I described above. We have not had one CLABSI since then in lines involving our team.

Jarvis: What percent or number of central lines (non-PICC) are inserted and/or maintained by clinicians other than your team and what is their CLABSI rate?

Harnage: Again, the VAT team is responsible for maintenance of all central lines, independent of what type of central line and which clinician inserted the line. Over the past two years, in conjunction with an aggressive early-recognition-of sepsis campaign, weve seen an increase in the placement of CVC catheters capable of SCVO2 monitoring, which measures central venous oxygen saturation. In part because of that, I cant give you an exact percentage of how many lines have been inserted by physicians instead of our team and how many insertions have been assisted by our team but not wholly inserted by them. Our focus has been immediately responding to those lines to make sure there was still complete bundle compliance. With all that said, the CLABSI rate for those lines is zero. All clinicians at Sutter use the same bundle our team does -- which indicates that the bundle is extremely effective.

Jarvis: Do you have plans to take over responsibility for all central lines (PICC and non-PICC) at your hospital in the future?

Harnage: What I can tell you is that we continue to expand our skills regarding central lines. Weve just completed a standardized procedure to have the VAT insert arterial lines in our critical care population. This procedure will be implemented by the end of the first quarter of 2013. The VAT is also exploring expanding our clinical skills to include insertion of internal jugular lines. One of the best things about working with our VAT is their deep, personal commitment to excellence. They are constantly seeking ways to leverage the knowledge of the medical staff to improve their own practices and the experience of the patient. This creates advantages for physicians and the institution, too, because it makes better use of physicians time. 

Jarvis: What definitions do you use for your PICC-associated BSIs and how do you or others do surveillance for those infections?

Harnage: Our infection preventionist Wendy D. Wiesehuegel, PhD(c), MS, BSN, RN, CIC, CR, directs us to the definitions from the CDCs National Healthcare Safety Network (NHSN) guidelines and protocols for the patient safety modules. As the infection preventionist (IP), she does the surveillance. She generates culture reports daily that are tailored to all NHSN modules for reporting: CLABSI, catheter-associated urinary tract infection (CAUTI), ventilator-associated pneumonia (VAP), clostridium difficile infection (CDI), and multi-drug resistant organism (MDRO).

Her methods for BSI detection include daily review of all culture reports according to the NHSN criteria for case-finding; review of input from various hospital clinicians on issues that may be related to BSIs; conducting weekly rounds; and reviewing patients who have high risk for BSIs.

She also runs raw lab data quarterly, including all completed blood cultures. The cultures are reviewed to determine whether they are primary BSIs that is, related to a central line catheter per the NHSN definition -- or whether some are secondary infections from a primary infection elsewhere in the body.

Jarvis: During this study period, have there been any changes in the surveillance methods used to detect PICC-associated BSIs (both the use of CDC definitions and the method for their detection)?

Harnage: No. We have consistently used the surveillance methodology I just described and also the same NHSN definitions.

Jarvis: For the years 2005 to 2012, what has been the number of PICC days each of those years?

Harnage:  We began tracking central line days in our ICUs in 2006.Prior to 2006, our IP tracked total number of discharges instead. Also, because our VAT is responsible for all central lines, we dont keep separate figures for PICC lines vs. all central lines.  With that said, the total number of CVC days for our ICUs are as follows:
2006 = 3,900
2007 = 3,947
2008 = 4,478
2009 = 3,562
2010 = 4,041
2011 = 3,492
2012 = 3,832

Jarvis:  Given the PICC-BSI rate in 2005 and before, can you tell us how many PICC-associated BSIs have been avoided with your program, what the cost savings from their prevention have been, how many lives have been saved and what the cost of your program vs. the cost savings have been over this period?

Harnage: I cant answer your question as precisely as you might like. For instance, I cant assign a precise program cost for our VAT team because it has other responsibilities besides managing central lines that dont relate entirely to CLABSI prevention. I can give you an estimate of how CLABSI reduction has reduced costs based on widely accepted figures.  We averaged 12 CLABSIs in the two years before we developed and implemented our central line bundle. Estimates of CLABSI treatment costs vary widely but one commonly cited range is $25,000 to $45,000. Multiplying that range by 12 CLABSIs, we have avoided $300,000-$540,000 in CLABSI treatment costs every year.  As for saved lives, another widely cited range is that 12 percent to 25 percent of CLABSIs are fatal. If we accept 12 CLABSIs as average for our pre-bundle years, we are now avoiding 1.44-3.0 deaths per year. Those are very rough numbers but it gives you an idea.

Jarvis: Obviously, standardization of practices is critical. Your PICC team is responsible for both the insertion and maintenance bundles. What role do the bedside clinicians (nurses and physicians) have in preventing these infections and how have you been able to standardize their practices?

Harnage: The VAT team drives the bundle, but all clinicians are expected to adhere to the bundle.  For the bedside staff, that includes the flushing protocol and the scrub-the-hub technique. As for standardization, we try to achieve it through education, data tracking, for example, we developed a flush audit tool to audit the flushing compliance of floor nurses, including demonstrated competency. New VAT members go through a several-week orientation program, which is thorough and covers all bundle components in detail. We also do in-service training on the elements of our central line bundle for the house staff, and we have reminder cards for all aspects of our bundle in reference binders at every nursing station. On top of the formal education and training, our VAT team is constantly educating and reminding clinicians to comply with the bundle as we do our daily rounds. Thats part of the ongoing responsibility for all VAT members.

Jarvis: Is your skin prep chlorhexidine alone or chlorhexidine with alcohol?
Harnage: We use chlorhexidine with alcohol.

Jarvis: What is your practice for disinfection of your needleless connector?
Harnage: We use the method commonly known as scrub the hub that is, before every line access, we scrub the connector hub with alcohol using downward pressure for a minimum of 15 seconds before accessing the line. As studies have shown, however, it can be challenging to fully implement this method especially when nurses are busy, which is always! We are constantly educating nurses to remind them of the importance of doing the method correctly. We support compliance and correct technique through supervision, as well.

Jarvis:  Do you require your home infusion company/companies to follow your protocol when patients with PICCs are discharged home?

Harnage: The short answer is no. For the patients in our own outpatient infusion center, we insert lines per our bundle but we cant monitor the lines daily once they go home. And in some areas, we dont want patients following our bundle exactly. We dont want patients flushing the lines every eight hours. If they access the line that often, there is a greater chance of contamination than if we did this in the hospital with all our training and the more controlled environment. They also dont have the same number of medications to administer as we do in the in-patient environment. For reasons like these, we give outpatients different instructions than we give our clinical personnel.

Jarvis:  Some have argued that elimination of PICC-associated BSIs is impossible.  How would you respond to that statement?

Harnage: Obviously, its not impossible because weve gone more than seven years without a CLABSI attributable to our team. Its more important to state what IS possible.  If your organization makes prevention a top priority with no excuses and no compromises, reduction and/or elimination of CLABSIs is possible. If administration supports the goal and provides the resources to focus on this issue, the elimination of CLABSIs is possible. I think our example shows what this means. Our administration addressed the difficulty of expecting every single clinician which includes more than 800 RNs and 200 MDs -- to be a content expert in this field. A focused, highly trained, and dedicated vascular access team that takes responsibility for central lines will increase upfront costs, but the outcomes result in not only clinical success but also financial success reducing length-of-stay, reducing readmissions, reducing healthcare-acquired infections, and promoting the safety of the patients. In short, if the organization values this work, it can expect good results.

Jarvis: Given the impressive success of your team, why wouldnt all the hospitals in your system and for that matter all the other hospitals in the region, California, and the country not copy this approach?

Harnage: The challenges facing hospitals across the country to reduce costs is an enormous pressure that every administration must face.  And the savings we have discussed are cost-avoidance savings. On the other hand, a dedicated VAT is an expenditure. Because the insertion of central lines, including both CVCs and PICCs, can be done by other hospital personnel such as ED, ICU and IR physicians, and the care of central lines is within the scope of the RN practice, theres a tendency to believe elimination of CLABSIs can be completed  with an if only approach.  If only everyone followed every protocol exactly every time.  If only there was adequate staff, with no registry or moonlighters. If only in the full scope of practice, no one needed to multi-task. But heres the problem with that approach. Picture your average ED or ICU physician. He or she has several very sick patients, and a patient presents with septic shock. The MD has to assess the patient, order appropriate labs, manage the multiple other symptoms, and insert a central line under full barrier precautions. The RN assisting the MD has other patients in the unit. While doing the procedure, an RN with another patient experiencing a heart attack comes into the room and shows the MD an EKG, requests a brief assessment, and asks for orders.  What just happened?  During a procedure that can take up to 30-plus minutes, the attempt to protect the sterility of the patient has been violated and the MDs concentration on the task at hand has been interrupted.

In our institution, the VAT RN preps the patient, completes the full barrier precautions, positions the patient, and assists the MD with identifying the vessel using ultrasound, and assists with the insertion. Once the line is inserted, the VAT RN secures the device, applies a CHG disc, dresses the insertion and flushes the line. The outcome: the physician needs only about 10 minutes to complete the procedure, full barrier precautions are maintained, and adherence to the bundle is complete, because that was the job of the VAT RN in this scenario. The VAT RN was not the clinician to diagnose the patient in the next room, hang the pressor, and measure the output or measure the temperature of the patient. The VAT RNs sole job is to assure the safe and timely insertion of the lifeline: the vascular access. This exact same scenario applies when it is the VAT RN inserting a PICC for a very ill patient.

How did we get here?  We started with the resources we had, our peripheral IV team. We then trained and educated them and partnered with administration and physicians champions to increase the scope of our practice. We developed the bundle after looking at our own practices, best practices, and evidence-based practices and conducted a thorough review of both practices and products.  And success begat success. When we saved MD time, improved patient access, improved patient satisfaction, and decreased CLABSIs, we were able to receive increased resources to do our job. In 2005, when we had 11 CRBSIs, we had two to three RNs working from 7 a.m. until 11 p.m., providing 16 to 24 hours of nursing care. Today we have five RNs providing 50 hours of care each day from 7 a.m. to 11 p.m.  We did not get to this place all at once.  We produced results and increased our benefit to the organization, and subsequently were given more resources. Our goal for 2013 is to become a 24-hour service.

Jarvis: What has been the biggest impediment or obstacle to the success of your team and how have you overcome that challenge?

Harnage: The biggest challenge isnt getting to zero, its sustaining zero. You are always challenged by the realities of the environment: turnover in physicians and nurses, higher patient acuities, and time pressures on practitioners. To sustain zero, you have to sustain education and training of the front-line staff. You also have to have commitment from administration to provide a focused, expert practitioner committed to assuring the right line, right patient, right diagnosis, and right therapy inserted in the safest manner. Finally, you have to have a bundle that is comprehensive and uncompromising. And we do!

Editor's note: Harnage acknowledges her supervisor Debbie Dix, RN, director of cancer services at Sutter Roseville, for her help in preparation of these responses.

NorthShore University HealthSystem
Patients at NorthShore University HealthSystem are now better protected against potentially deadly infections, according to a newly published study. The research found that use of a disinfection cap on catheters and tubing reduced infection of central intravenous catheters by more than half. The study, which involved all four hospitals in the NorthShore system, was published in the January 2013 issue of the American Journal of Infection Control.

Because of the study results, use of the disinfection cap has been adopted as a standard of practice at NorthShore for all central IV catheters. 

Most hospitals have a hard time assuring proper disinfection of IV needleless connectors, says Marc-Oliver Wright, MT (ASCP), MS, CIC, the studys lead author and corporate director of infection control for NorthShore. This is a real concern because contaminated connectors are a major source of central line associated bloodstream infections, or CLABSIs. We found that a disinfection cap was a very effective solution to the widely recognized shortcomings of the standard method for disinfecting IV connectors.

Researchers also assessed the cost-effectiveness of the cap and concluded that its use resulted in net cost savings.

The study was undertaken because the standard protocol for manually disinfecting connector hubs, called scrub the hub, often fails because time-pressed nurses are not always able to properly scrub IV connectors with alcohol.

CLABSIs are a serious and often fatal form of infection that many hospitals have found difficult to control. The scrub the hub method of manual disinfection, with its inherent challenges, is often cited as a potential cause when a hospitals CLABSI rate is high.

The study was designed to determine whether a disinfection cap overcame problems with manual disinfection. The cap is a plastic device that twists onto the threads of the IV connector. Attaching it to the connector compresses a medical-grade foam pad inside the cap that releases isopropyl alcohol (IPA) onto the connector hub to disinfect it. The cap is left in place between IV line accesses, and its retention seal keeps the connector bathed in alcohol while protecting it from contamination by external sources.

An IV connector is a device that connects an IV catheter to tubing and helps deliver medications and nutrition. Incomplete disinfection of a connector can allow bacteria to enter the connectors fluid pathway and cause an infection.

Disinfection caps were invented because of the problems with scrub the hub, Wright says. Theoretically, they should compensate for those problems and provide additional protection for patients, as well. But you never know if something will work as intended until you do the research. Our study design showed in multiple ways that the cap was associated with fewer infections.

Q&A with Marc-Oliver Wright, MT (ASCP), MS, CIC
We discussed the secrets to Northshore's success with Marc-Oliver Wright, MT (ASCP), MS, CIC, corporate director of infection control for NorthShore University HealthSystem

Q: You and your colleagues were hoping to reduce catheter-related bloodstream infections. Why is that such an important issue for clinicians and why did you focus there?

A: To be completely honest, at the onset of the study we thought that the likelihood that the device we were evaluating would reduce infections was really slim. Thankfully, CLABSIs had already declined to very low rates (both nationally and within our own organization) and the possibility of observing a reduction that approached statistical significance was not anticipated. Our primary objective was demonstrating efficacy of the device itself in reducing or eliminating intralumninal contamination from the catheter hub. In other words, we sought to answer the  question does the device do what the manufacturer claims? The evaluation of infections was secondary; sometimes people assume that intralumninal contamination from the hub immediately/always causes infection, but that just isnt true and we dont know how much is too much before someone develops a line infection. Dr. Leonard Mermel articulated the primary causes of line infections in a really nice viewpoints article for Clinical Infectious Diseases in 2011 where he made a clear outline of why our traditional five causes of line infection can really be narrowed down to two; the skin at the insertion site and the catheter hubs. With our, bundles, checklists and our antiseptic soaked sponges at the insertion site, we had addressed the former, but the latter was left to scrub the hub campaigns which we didnt think were any more long lasting than other awareness rousing campaigns.

Q: Your study involved using a disinfection cap on the needleless valves at the end of the catheters and tubing as a way of compensating for lapses in the manual method of disinfection and better protecting patients against bloodstream infections. Describe the disinfection cap, including how it works and how it is supposed to be used.

A: The disinfecting cap itself really reminds me of a wire nut or connecting end cap used in simple home electrical work which you use to join two wires together. Its basically a small, bright orange, plastic cap with interior threading that allows the user to twist the cap on to the unused needleless valve. Inside the cap is a small sponge saturated in 70 percent alcohol. When a needleless valve is not in use, this cap is secured on to the valve and left in place until the next time the valve needs to be accessed (up to four days) when the cap is removed and discarded. After you access the needleless valve you take a new disinfecting cap and apply it to the valve and walk away. Its single use and acts as a physical and chemical barrier between the patients intralumninal space and the patients environment.

Q: Would you summarize the methodology of your study and its results?

A: This was a quasi-experimental (baseline-intervention-baseline) study at four Chicago area hospitals of adult patients with central lines divided into P1 (baseline), when the standard scrub was used; P2, when the cap was used on all central lines; and P3, when standard disinfection was reinstituted. Adults with peripherally inserted central catheters (PICCs) inserted during hospitalization having five or more consecutive line-days were enrolled, and the vascular access nurse collected 1.5 mL of blood as the first drawn specimen from each lumen not in use and quantitatively cultured. In addition, we looked at house-wide central-line associated bloodstream infection (CLABSI) rates for all patients with central lines (not just PICCs) and used catheter-associated urinary tract infections (CAUTI) as a control measure.

Contamination rates dropped by more than half (12.7 percent vs. 5.5 percent) and bounced back to baseline (12.0) when the cap was removed and we returned to scrubbing the hub. The number of bacteria recovered (because we did quantitative cultures) dropped by 75 percent. Lastly, CLABSIs dropped by half, from an arguably low baseline of 1.43 per 1,000 line days to 0.69 per 1,000 line days. All of the above were statistically significant. When we removed the cap from one hospital, infection rates did increase to near baseline levels, but this increase was not statistically significant. The rates of CAUTI, as our control measure, did not correlate with the changes we saw with CLABSIs, suggesting that some alternative infection prevention initiative or performance improvement was not at work diminishing our infection rates, further reinforcing what we believe to be an effect of the disinfection cap.

Q: How would you assess the strength of your study?

A: I was trained as an epidemiologist; so to me, good study design is really the most important thing. Its also really hard to do in the setting of infection prevention efforts because most of the time, we are responding to an existing problem and HAIs, despite their prominence in healthcare today, are thankfully becoming increasingly rare, making it harder and harder to conduct studies that end up having statistically significant findings.
This study was done in multiple hospitals, across multiple units, including non-critical care areas, which I think makes the findings more generalizable. We included a control measure which demonstrated that the findings were more likely to be due to the intervention rather than potential general infection prevention improvements. We include an assessment of cost, which many might overlook. Practically speaking, as infection preventionists we have to develop the skill set for making the business case. A better way to do this study would have been a randomized controlled clinical trial, but operationally, that would have strained our resources to a breaking point. Maybe next time.

Q: In the discussion section of your study, you state that the two most modifiable causal pathways for preventing bloodstream infections include one extraluminal source i.e. skin organisms and one intraluminal source: catheter hub contamination. Obviously, disinfection cap use can address the intraluminal source but does it have any relevance for the extraluminal one?

A: I dont think so. But then again, weve already done a lot to address the extraluminal source. Chlorhexidine gluconate (CHG) for skin prep, CHG sponges at the insertion site, all the elements of the checklist for insertion; all of these are aimed at reducing the potential for infection from extraluminal sources.

Q: Using a device for catheter needleless valve disinfection obviously adds some upfront cost. You performed a business analysis on the cost-effectiveness of disinfection cap use. What were the factors you measured and what did the analysis show?

A: Originally, I did the business case using hard numbers from internal estimates of how much my organization charges for a bed, how much we spend on antibiotics to treat a CLABSI and follow-up lab results etc and ended up with a final dollar amount the organization saved by using this product. I then shared this method with a mentor who knows more about business case development and healthcare economics than anyone I know (Eli Perencevich, MD, of the University of Iowa) who suggested the Ward approach I ended up using in the paper. What I like about this is that it gets at the human element and leaves the reader to decide if the cost is worth it. The product (cap) costs were a direct cost reported from our purchasing department annualized out over a year of central line days. We knew our baseline infection rate would be cut by half, leading us to determine we would avoid 21 CLABSIs in using the product. The length of stay estimate was more conservative than previous reports, but the methods used by the researchers in estimating the true attributable length of stay for a CLABSI were more reliable so we went with that estimate of 2.7 days. The 20 percent mortality estimate is a common mortality rate used in the literature. Lastly, we used our own average length of stay 4.5 days to back calculate the number of additional admissions we could afford if we avoided the 21 CLABSIs. The final result is that for a cost of $2.07 per catheterized patient per day, we avoid 21 infections and 4 deaths and make enough beds available for 13 new admissions. Notice, we tell you everything, how much we spend on the product, how many infections are avoided, lives saved, bed days made available; what we dont tell you is a final dollar amount. As an inherent cynic, I like using the most conservative estimates possible and leaving the reader to decide if $2.07 is too much, too little  or just right for what our benefit was. If we had used the literature based estimate of $30,000 per CLABSI, our organizations one-year savings would be $390,617. I dont believe that number myself and I dont believe anyone else should either. 

Q: Needleless valve disinfection is not something that is usually addressed in conventional central line bundles. First, do you think it should be and second, do you think use of a disinfection cap should be a recommended practice? 

A: First, I can say that the use of a disinfecting cap is now standard operating procedure at my organization. The reaction from staff in using it as well as the results of the study convinced us that it was the right thing to do. That being said, I dont know if a single study even in a multifacility setting should be considered enough to change national guidelines. I know when I read the paper by Timsit and colleagues where they found a 60 percent reduction  in CLABSIs in using a CHG sponge at the insertion site I was really excited and brought the new product in immediately. Instead of a 60 percent drop, I saw a 22 percent drop. That was enough to justify the additional cost, but not what I had expected. Just because we are down to two primary causes of CLABSI, it doesnt mean that that spilt is 50-50 between the two or uniform across all settings. Notice, this study was only done on adult patients. An evaluation of its potential for pediatrics is really warranted. If additional studies show a decrease in CLABSIs when using a disinfecting cap it would make sense for their use to become a recommended practice. I personally believe that bundles, when they are appropriate for use, should follow guideline recommendations.

Q: Is use of a disinfection cap a good fit with standard nursing practice?

A: Actually, yes it is. As opposed to scrubbing the hub, the action related to compliance with this device (putting a new one on) is done after the (mentally) higher priority task (administering medication or collecting lab specimens) is completed. If our human factors experts could (and have) argued that many infection prevention initiatives (e.g. hand hygiene) are interrupters to what is perceived as the higher priority task (patient care) it makes sense to design new approaches to infection prevention that avoid interruption in order to increase compliance, improve effectiveness and ultimately reduce patient harm. Our nursing staff seem to be very happy with the device being standard practice in our organization and were happy with where thats taken us in preventing avoidable patient harm.