By Marilyn Hanchett, RN, MA, CIC
Therapies administered via vascular access catheters are a mainstay of both acute and chronic medical treatment. In fact when all types of vascular catheters are combined, the category may well represent the largest type of invasive device used in healthcare. However their ubiquitous presence, ironically, makes it very difficult to measure with any degree of statistical precision their associated device utilization rates. The use of peripheral catheters (PIVs), for example, is so widespread in the United States that their use is obscured among other procedure codes, rather being coded separately, and recorded uniquely only in individual patient records. Aggregate data is absent in any administrative datasets. In general the frequency of catheter use must be inferred from the number of units sold by their various manufacturers. Today, this number reportedly exceeds 300 million but is at best an estimate.
The notable exception is the use of central catheters, including those inserted peripherally. The risk of serious infectious complications supports careful monitoring of these “central lines,” specific CDC surveillance definitions and guidelines, and the maintenance of the CDC's NHSN database for trending. The attention focused on central catheters has continued for years so that now in the United States it is an essential component of both the HHS HAI National Plan longitudinal improvement measures as well as a basis for CMS hospital inpatient reimbursement for individuals whose care requires the use of this type of device.
The intense scrutiny of central catheters, often to the exclusion of all other types of intravenous devices, is based on risk assessment. Multiple studies document the complications linked to their use, including catheter associated bacteremia.(1-2) Far fewer studies describe the complications of peripheral devices. An often cited review of 2,000 prospective studies by Maki in 2006 reports the infection rate for PIVs at 0.1 percent or 0.5 per 1,000 intravascular device days.(3)
Subsequent studies have reported both lower and higher rates. However sample sizes vary and the use of denominators is inconsistent. For example, research have used dominators such as patient days (or occupied bed days) and catheter hours (rather than days). This significant variation in definitions and statistical methods make comparative analysis virtually impossible. A current and valid estimate for all intravascular associated infections, and specifically for those excluding central catheters, remains unknown.
This means that any general assumptions underlying the organization of surveillance activities addressing intravascular therapies are somewhat tenuous. For example, there is no NHSN device module for non-central catheters. Facilities may assume that infections linked to other vascular access devices may or may not be captured as an underlying component of the MRSA Lab ID bacteremia reporting process. The NHSN MRSA bacteremia LabID event could possibly be linked to device use (thus validating the assumption that those infections are in fact identified, at least indirectly) but the reporting process does not collect that level of detail. Therefore, a 2013 CDC study of 2,885 hospital-onset (HO) MRSA bacteremia LabID events based on more than 44 million patient days identified three factors in predicting HO MRSA bacteremia Lab ID events: medical school affiliation, facility bed size and community onset MRSA bacteremia prevalence rate.(4) This study provides insight into potential demographic, rather than patient care, risks. The role of non-central catheters as a contributing factor remains unknown.
Not only must venous access devices other than central line be considered, but bacteremia associated with organisms other than MRSA cannot be ruled out. Scrutiny limited to MRSA bacteremia may unintentionally lead to a false sense of surveillance security, since increasing rates of organisms other than MRSA, both community and hospital associated, as well as increasingly resistant strains have been reported.(5)
Assumptions may be the greatest obstacle to integrating all types of vascular access catheters into a surveillance system. Perhaps the most dangerous assumption is that an absence of known PIV associated infections offers sufficient justification for not conducting broader investigations. It is faulty logic to infer that a lack of targeted surveillance equates to an absence of potential infectious risks and/or actual infections. It is also tenuous, given the lack and strength of evidence, to conclude that the financial burden of applying hospital resources to investigate a high volume, presumably low risk device (such as a peripheral catheter) is not cost effective, especially considering the pressure of other types of surveillance needs and public reporting requirements that facilities must meet. This position attempts to preempt expanded surveillance based on the assumption that the return on the investment of already limited staff resources for the identification of new infection cases will not be justified either by the number of new infections reported and/or clarification of otherwise unrecognized risks. This assumption may actually be correct, but only after the facility’s surveillance has demonstrated that the devices under consideration and utilization rates truly can be classified as low risk.
Few, if any, professionals responsible for healthcare surveillance will be anxious to add to the existing data burden. Therefore a facility considering expanded surveillance approach to intravascular devices should consider the following factors.
1. Limited Evidence
Published studies on catheters other than central lines focus predominantly on noninfectious complications, e.g., phlebitis, occlusions, infiltrations and extravasation. While the role of phlebitis as a precursor to infection has been suggested,(6) there is insufficient evidence to reliably extrapolate bacteremic risks from the phlebitis rate even if hospital specific phlebitis rates are routinely tracked.
2. Lack of Standard Definition
The limitation of evidence leads directly to the absence of a standard infection case definition. Different types of peripheral and other non-central catheters, different dwell times, different types of infusions, frequency of catheter use, different securement methods, as well as a wide variation in patient variables (age, availability of peripheral access, skin condition, and cognitive function) are all variables that make any consensus on standard criteria extremely challenging. In the absence of a standard definition, the facility must be prepared to develop their own on available scientific evidence and its facility specific data or case studies. The facility must also determine if patients who self-administer their therapies under hospital or provider supervision will be part of the definition.
3. Need for a Time Sensitive Denominator
The mean dwell time for PIVs is unknown, but these devices are widely used for short term, intermittent therapies where the infusate does not require administered in a larger vein. For that reason the accepted use of device days may be irrelevant for this type of monitoring. The same dilemma exists if patient days are used, especially when including outpatients or any encounter where the patient is not admitted for more than 24 hours.
4. Difficulty of Patient Tracking and Follow Up
Monitoring patients whose devices have most likely been removed and who may show symptoms of infection several days after their infusion presents a serious obstacle to surveillance. How and when post device use tracking will occur is therefore an essential component of planning for any level of expanded monitoring.
If a facility chooses to expand their intravascular catheter surveillance system beyond central lines, it is essential that the process include all of the stakeholders. No one individual or department can be solely responsible for a far reaching surveillance program. The engagement and assignment of tasks, data analysis and reporting responsibilities must be carefully planned and integrated into existing workflows. It is only through this approach that the infection prevention community will eventually better understand the risks associated with the spectrum of intravascular catheter use and be able to design effective and efficient surveillance systems to help mitigate those risks.
Marilyn Hanchett, RN, MA, CIC, is the infection preventionist at Frederick Memorial Hospital in Frederick, Md.
1. Stevens V, Geiger K, Concannon C, et al. Inpatient costs, mortality and 30 day readmission in patients with central line associated blood stream infections. Clin Microbiol Infect 2014 May; 20(5)0318-24.
2. Wisplinghoff H, Bischoff T, Tallent SM et al. Nosocomial blood stream infections in US hospital: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004 Aug; 39(3):309-17.
3. Maki DG, Kluger DM, Crinch CJ. The risk of bloodstream infection in adults with different intravascular devices: a systemic review of 200 published prospective studies. Mayo Clin Proc 2006/ Sept; 81(9):1159-71.
4. Dudeck MA, Weiner LM, Malpiedi PJ et al. Risk adjustment for healthcare facility-onset C difficile and MRSA bacteremia laboratory identified event reporting in NHSN. Published March 12, 2013. http://www.cdc/gov/nhsn/pdfs/mrsa-cdi/Risk Adjustment-MRSA-CDI.pdf.
5. D’Agate EM, Horn MA, Ruan S, et al. Efficacy of infection control interventions in reducing the spread of multidrug-resistant organisms in the hospital setting. PLoS One 2012;7(2):e30170.
6. Gonzales Lopez JL, Arribi Vilela A. Fernandez del Palacio E et al. Indwell times, complications and costs of open vs closed safety catheters: a randomized study. J Hosp Infection 2014 Feb;86(2):117-26