Nathaniel R. DeFelice, MD, and Jennifer A. Best, MD

Dr. DeFelice is Clinical Instructor, Division of General Internal Medicine, Harborview Medical Center, University of Washington, Seattle, and Dr. Best is Associate Professor, University of Washington School of Medicine, Seattle

Drs. DeFelice and Best report no financial relationships in this field of study.

SOURCE: Chopra V, et al. Do clinicians know which of their patients have central venous catheters?
A multicenter observational study. Ann Intern Med 2014;161:562-567.

Central venous catheters (CVCs) are essential to providing optimal care to many hospitalized patients. While traditionally limited to critical care patients with triple-lumen catheters inserted into the subclavian, internal jugular or femoral veins, the expanding use of peripherally inserted central catheters (PICCs) has created an environment where most CVCs are now found in non-ICU patients. The increase in their use throughout the hospital is not surprising as they offer many benefits to patients, physicians, phlebotomists and nurses ranging from pain-free blood draws to hemodynamic measurements guiding clinician decision-making. However, even as CVCs (especially PICCs) become more common on general medicine floors, we are becoming increasingly aware of their associated risks. These risks include costly and potentially life-threatening central line-associated bloodstream infection (CLABSI) and venous thromboembolism (VTE). While we know that the risk of these complications increases with time, a growing body of evidence has demonstrated that the inappropriately prolonged use of CVCs is not uncommon.

Suspecting that provider unawareness of the presence of CVCs in hospitalized patients contributes to unnecessarily prolonged use, Chopra and colleagues conducted an elegant multicenter, cross-sectional study to assess clinician awareness of their patients’ CVCs. The authors randomly selected hospitalized patients and their responsible providers at 3 academic medical centers in the U.S. Between April 2012 and September 2013 the investigators conducted patient interviews and performed focused exams before morning rounds where they noted presence or absence of a PICC or non-tunneled triple-lumen catheter placed in the neck, chest or groin (other types of specialty catheters were excluded). Then, after the team rounded on their patients, the providers (n=990) were asked to identify which of their patients had a CVC.

The investigators hypothesized that clinicians who are most “proximal” (such as interns or hospitalists) would be more likely to know which of their patients have CVCs. They also postulated that providers such as critical care physicians who often insert lines or oncologists who often consciously deliberate on the choice of vascular access would be more likely to remember that their patient has a CVC. When appropriate, they used Chi-square tests to compare differences between interns, residents, general medicine teaching attendings, and hospitalists. Additionally, differences between critical care, subspecialty and general medicine services were explored.

Across all sites, CVCs were present in 21% (209/990) of patients, the majority (60.3% [126/209]) of which were PICCs. In composite, providers believed that a CVC was absent when it was present in 21.2% (90/425) of cases, while another 5.6% erroneously believed a CVC was present and another 9.1% admitted they were unaware of the presence or absence of a CVC.

Contrary to the study’s initial hypothesis, investigators identified hospitalists as the worst offenders — unaware of CVC presence 30.5% of the time (18 of 59). Interns were better, but still almost 1 and 5 of them were unaware of the presence of a CVC in their patient despite being the most likely to write orders (19.1% [22 of 115]). At 13.8%, residents were not significantly better than interns (P=0.027). General medicine teaching attendings and hospitalists were significantly more often unaware of the presence of CVCs when compared to interns, residents and non-physician providers (27.3% vs. 16.4%; P=0.006). As expected, critical care providers were much less likely to be unaware of CVC presence than general medicine or hospitalist physicians (12.6% vs. 26.2%; P=0.003). Subspecialty providers who were felt to be more likely to deliberate about access were not significantly better than their general medicine and hospitalists colleagues at identifying which of their patients had CVCs (22.5% vs. 26.2% were unaware of CVC presence; P=0.48).

The study had several limitations including determination of provider awareness at only a single point in time. Additionally, it remains to be proven whether lack of awareness correlates with patient outcomes or length of stay.

In their discussion, the authors suggest that the larger patient volumes of hospitalists compared to their ICU and house staff colleagues combined with the absence of a culture of paying attention to these lines outside the ICU may explain why hospitalists lacked relative awareness of their patients’ CVCs. A shift in hospital culture and systems to ensure that providers are more mindful of the presence of CVCs was encouraged, but further research is needed to explore what types of interventions are effective and if knowledge of CVCs improves outcomes in patients.

In conclusion, providers and particularly hospitalists are frequently unaware of the presence of CVCs in their patients, which could lead to inadvertent prolonged use and potentially dangerous complications. Further research is warranted to determine the impact provider unawareness of CVCs has on patients and what interventions best address this awareness deficiency.