Variability in Barrier Precautions Use Raises Red Flag About Arterial Catheter Infection Risk

Variability in Barrier Precautions Use Raises Red Flag About Arterial Catheter Infection Risk

An expert in medical device-related infections is trying to raise awareness of the variability that exists in following guidelines designed to help prevent catheter-associated infections, particularly in arterial catheters (ACs) used in intensive care units (ICUs) and operating rooms (ORs). Leonard A. Mermel, DO, ScM, medical director of the epidemiology and infection control department at Rhode Island Hospital, and his colleagues recently published results of a survey (Cohen, et al. 2015)  showing significant variability regarding how clinicians manage arterial catheters in ICU patients and that these practices may increase risk of infection.

The Rhode Island Hospital infectious diseases team includes (from left to right): Jason T. Machan, PhD; Andrew T. Levinson, MD; Leonard A. Mermel, DO, ScM; Gerardo P. Carino, MD, PhD;  and Stephanie N. Lueckel, MD. Not pictured are Dorothy Skierkowski, MA; Daithi S. Heffernan, MD; David M.Cohen, MD, and Jeffery Mazer, MD. Photo courtesy of William Murphy, Rhode Island Hospital

By Kelly M. Pyrek

An expert in medical device-related infections is trying to raise awareness of the variability that exists in following guidelines designed to help prevent catheter-associated infections, particularly in arterial catheters (ACs) used in intensive care units (ICUs) and operating rooms (ORs). Leonard A. Mermel, DO, ScM, medical director of the epidemiology and infection control department at Rhode Island Hospital, and his colleagues recently published results of a survey (Cohen, et al. 2015)  showing significant variability regarding how clinicians manage arterial catheters in ICU patients and that these practices may increase risk of infection.

The arterial catheter is commonly used in intensive care medicine and anesthesia to monitor the blood pressure and sample blood gasses directly from the artery. Approximately 8 million arterial catheters are placed in U.S. hospitals every year. This type of catheter is a potential source of bloodstream infections because it provides a direct, indwelling, and frequently accessed pathway between the skin and bloodstream.

Fewer than half of those surveyed complied with current Centers for Disease Control and Prevention (CDC) infection prevention guidelines for arterial catheter insertions.

"Barrier precautions are employed inconsistently by critical care clinicians across the nation, and such individuals underestimate the infection risks posed by arterial catheters," says Mermel, co-author of the study. "Every effort should be made to prevent such infections since they lead to increased cost, length of stay and morbidity."

Mermel says that in his discussions with some other physicians, he sensed that they did not perceive risk of infection with arterial catheters to be significant. "I think many physicians have focused a lot of their risk-reduction efforts and attention on central venous catheters, because they see arterial catheters as being one of those short-term devices that has a very low risk of infection. It concerned me that the perceived risk was not as high as it really is. In chatting with my colleagues we thought it was important to query people using arterial lines around the country to get a sense of what they think the risk is.   Based on the published literature -- the risk is half to two-thirds of that of a central line when it comes to bloodstream infections but we showed that the perceived risk is actually much lower. If the perceived risk is much lower, then precautions used to prevent those infections may not be as scrupulous as needed."

As Cohen, et al. (2015) note, "There is a growing body of evidence that ACs pose an infectious risk that is comparable to CVCs. Recent studies have shown that the occurrence rate of BSIs (new infections per 1,000 catheter-days) associated with ACs is 0.9–3.4/1,000 catheter-days, which corresponds to 40 percent to 90 percent of the occurrence rate of BSIs associated with CVCs."

Prior to 2011, there were no guidelines specifying which barrier precautions should be used during AC insertion.  That year, the CDC published recommendations specifying that clinicians don sterile gloves, a surgical cap and surgical mask, and use a small sterile drape when these catheters are inserted into the artery of a patient. Of the 1,265 study respondents, Cohen, et al. (2015) report that only 44 percent said they were using the CDC-recommended barrier precautions during insertion and only 15 percent reported using full barrier protections.

"There appears to be a significant deviation from clinical guidelines regarding a very commonly performed procedure in critically ill patients," says Andrew Levinson, MD, MPH, corresponding author. "Bloodstream infections are largely preventable, and if the survey results mirror the clinical practice in the U.S., there's work to be done in reducing risk of such infections."

Mermel and his colleagues note that to date, there have been no large-scale studies evaluating which aseptic techniques are actually used by clinicians during AC insertion in clinical practice and that it is unknown whether critical care clinicians in the United States are aware of or in compliance with the current CDC guidelines for AC insertion. The researchers hypothesized that significant practice variability exists with regard to infection prevention techniques used during AC insertion in the ICU setting, and that clinicians underestimate the infectious risks posed by ACs.

Survey respondents were members of the Society of Critical Care Medicine, specifically attending physicians, fellows, residents, nurse practitioners, physician assistants, registered nurses and respiratory therapists. The researchers report that beyond the level of training which was significantly related to self-reported compliance with CDC guidelines during AC insertion, there were no statistically significant variations in self-reported compliance with CDC guidelines with respect to specialty, critical care training, hospital size, ICU size, frequency of AC insertion, status as a teaching hospital, geographic region, or perceived risk of infection associated with ACs.

Cohen, et al. (2015) report that clinicians in this anonymous survey "significantly underestimated infectious risk posed by ACs. The mean and median estimates of the relative risk of BSIs associated with ACs when compared with CVCs (in terms of occurrence rate) were 0.15 and 0.05, respectively. Using the reference occurrence rate provided to participants of 2.2 BSIs per 1,000 catheter-days associated with CVCs, these correspond to a mean estimated occurrence rate of BSIs associated with ACs of 0.3/1,000 catheter-days and a median estimated occurrence rate of 0.1/1,000 catheter-days."

According to the survey, 15 percent of respondents reported use of full barrier precautions during AC insertion, consisting of hand hygiene, cutaneous antisepsis using alcoholic chlorhexidine, sterile gloves, sterile gown, surgical cap, surgical mask, and full body sterile drape. As Cohen, et al. (2015) emphasize, "One possible explanation for this poor compliance with CDC guidelines is an apparent misconception about the infetious risk posed by ACs. On average, clinicians in this survey underestimated the risk of infection associated with ACs by a factor of 3, and the majority of clinicians surveyed (based on the median) underestimated the risk of infection by a factor of 10 or greater."

A potential explanation of the poor compliance with recommended precautions, according to the researchers, is that clinicians may be simply unaware of the updated CDC guidelines published in 2011. They note, "Clinicians in this study were not asked whether they were aware that CDC guidelines existed due, in part, to concerns that such a question might alter their responses. Subsequently, the contribution of lack of knowledge of current guidelines by clinicians in this study is unknown. Interestingly, a multicenter survey of CVC insertion policies in 2002, prior to the implementation phase of the Surviving Sepsis campaign, revealed only 28 percent compliance with then-current CDC guidelines. This study highlights the potential of large-scale educational initiatives to propel change." They add, "The reasons for poor compliance may be multi-factorial, including clinicians’ unawareness of current guidelines, misperception of infectious risk, disbelief in the efficacy of current guidelines, or other motivating factors yet to be determined. Regardless of the root cause, if the results of our survey are representative of clinical practice in the United States, then our current level of compliance with CDC guidelines for AC insertion represents a missed opportunity to prevent BSIs in the intensive care setting."

Mermel acknowledges that time constraints or the lack of access to PPE and drapes may be compounding the problem. "Do institutions have an arterial line insertion kit, as they probably all do for central lines?  My guess is that they may not," Mermel says. "And if they don't, then they are less likely to be consistently compliant with best practices. Also, if there is not a perceived risk, then that may decrease the motivation for using barrier precautions."

What's interesting is that 39 percent of clinicians surveyed indicated that they would support mandatory use of full barrier precautions for AC insertion. As Cohen, et al. (2015) note, "These findings indicate that if the CDC guidelines for AC insertion are further revised to require full barrier precautions, such changes may be met by considerable resistance. The major limiting factor in adoption of full barrier precautions by clinicians in this study was use of the full body sterile drape, which was used by only 20 percent of respondents. To date, no randomized controlled trials have compared the efficacy of different drape sizes in decreasing the incidence of BSIs associated with ACs. However, it seems unlikely that a small area drape can ensure equivalent sterility of both the proceduralist and procedural field, especially when ultrasound guidance is used for AC insertion. Adequately powered, prospective, randomized studies are needed to develop an evidence-based infection prevention bundle for AC insertion."

"We have an arterial line kit at our hospital," confirms Mermel, that isfully compliant with the CDC recommendations. "However, I don't know how many other institutions have similar kits. They are convenient because all of the necessary components are immediately accessible, and the clinician putting in the catheter doesn't have to go to the clean utility room for supplies. The place to start is to educate regarding the risk so that we can make it easy for them to do the right thing. That's the challenge of infection control, certainly."

Mermel says the arterial catheter insertion kit was assembled some years ago, with its components agreed upon by extrapolating from the experiences of central line insertions. "It definitely required a culture shift whereby we created the kit and it became a hospital standard. We have set the expectation that kits will be accessible and will be used, and the kit components are established through infection control and user input, so that those actually doing the procedure are satisfied that the kit meets their needs."

Cohen, et al. (2015) say that in light of recent studies suggesting that the risk of bloodstream infections associated with ACs is comparable to CVCs, "a compelling argument can be made that the insertion of ACs should require the same barrier precautions as CVCs." The researchers add, "In addition, a propensity-matched cohort study of AC use in mechanically ventilated ICU patients failed to identify a subset of patients in which ACs offered a mortality benefit. Secondary analysis found that AC use was associated with increased mortality in patients receiving vasopressors. Although not definitive, these findings, if corroborated by randomized controlled trials, may in part be attributable to AC-related infections. However, compared with the infection prevention bundle used for CVCs, the CDC-recommended precautions for ACs are less stringent and less evidence-based. Of the three studies cited in support of the CDC’s recommendation of limited barrier precautions during AC insertion, two are single-arm studies in which more stringent precautions were used than those recommended by the CDC. The third supporting study is a small, randomized controlled trial that compared limited barrier precautions with full barrier precautions for the insertion of ACs (25). This study showed no difference in colonization and a nonsignificant decrease in AC-related infections in the full barrier precaution arm (relative risk = 0.4; p = 0.11). However, with only 272 participants, this trial was underpowered; the authors of the study estimated that 2,200 randomized participants would have been required to detect a 50 percent reduction in AC-related BSIs. Of note, this is the only randomized controlled trial that has evaluated the efficacy of full barrier precautions for the insertion of ACs."

Mermel says there is a significant need to raise awareness of the risk of serious infections from arterial catheters, especially since most have an extraluminal source of infection, that is, coming from the skin predominantly. "This is based on data from research at the University of Wisconsin," Mermel says. "If that's the case, then barrier precautions become even more important and we need future to confirm this. My preference would be to use full barrier precautions and play it safe rather than having minimal precautions, particularly knowing that the predominant source is the skin. And now with some clinicians using ultrasound and other modalities, you need a pretty big sterile field."

The current variability in practice doesn't surprise Mermel, who acknowledges that guidelines and recommendations can only go so far to help inform procedures, and that local needs and situations quickly alter implementation. "A guideline is just a guideline and doesn't supplant clinical judgment," Mermel says. "In a JAMA article I co-wrote with Dave Classen, we noted that guidelines are simply a framework to work from -- there will inevitably be some differences in practice based on patient-to-patient variability. Our arterial catheter survey suggests that the variability is significant and that's due, in part, to differences in perceived risk of infection."

What also concerns Mermel is how arterial catheters are treated as "spigots," prompting him to write an editorial, "Arterial Catheters are Not Risk-Free Spigots," that was published in Critical Care Medicine (February 2008; Vol. 36, No. 2; pp 620-622). "In ICUs, ACs are easy access for frequent blood drawing," Mermel says. "Clinicians must ask themselves, are these catheters needed, and what are the precautions required.  In terms of maintenance I think they are used very frequently just because they represent immediate access, and sometimes clinicians are loathe to remove them because it's direct access to the bloodstream and they can draw blood frequently. We know from arterial catheter pathogenesis studies that many of the infections, about two-thirds, come from the skin rather than the catheter hubs. So it's important and clinicians should be required to document daily in the medical record if  an AC or another invasive device is still required for patient care. ACs are heavily accessed in the ICU and the driver of that is easy access to the bloodstream, lots of bloodwork being done, and caregivers become dependent on these catheters. But do they really need all those labs every day? Excessive blood drawing is something we are studying right now -- in one of our posters we called it 'vampire medicine.' It's convenient for bloodwork, not all of which is needed."

Mermel acknowledges that many institutions have made great strides in reducing risk from central lines in the ICU, but much more work needs to be done. "We need to take all of the infection prevention efforts that we have conducted over the last decade in the ICU for central lines and start transitioning to devices used in other healthcare settings such as the general wards and where patients have central lines and  peripheral IVs. –  The latter is another underappreciated infectious risk to patients. If people think it's just a little IV and can't cause a bloodstream infection, they need to think again -- the bloodstream is interconnected so be it a central vein or a peripheral vein, if you get Staph migrating down the catheter it will eventually cause a bacteremia."

The take-home message for clinicians and infection preventionists alike is to refocus device-related infection prevention efforts, according to Mermel. "Now that we have made great strides with central lines in ICUs, we need to implement some of our prevention strategies with central lines outside of the ICU," he says. "We need to pay more attention to peripheral lines, and back in the ICU, we need to look harder at arterial catheters with regard to considering an insertion kit and making it easier for clinicians to do the right thing. We must raise awareness with regard to device need -- we must work within the cultures of the ICUs and the ORs to see if we can push clinicians to remove the arterial catheters and other devices if they are not required for patient care."

References and Resources:

Centers for Disease Control and Prevention (CDC). 2011 Guidelines for the Prevention of Intravascular Catheter-Related Infections. Accessi-ble at:

Cohen DM, Carino GP, Heffernan DS, Lueckel SN, Mazer J,  Skierkowski D,  Machan JT, Mermel LA and Levinson AT. Arterial Catheter Use in the ICU: A National Survey of Antiseptic Technique and Perceived Infectious Risk. Crit Care Med. Aug. 7, 2015 epub ahead of print.

Marschall J, Mermel LA, et al. Strategies to Prevent Central Line–Associated Bloodstream Infections in Acute Care Hospitals: 2014 Update. In-fect Control Hosp Epidem. Vol. 35, No. 7, pp. 753-771. July 2014. Accessible at:


Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.