Prevention of Thrombophlebitis in Peripheral Intravenous Catheters: The Butterfly Effect

In this issue of Anesthesia & Analgesia, Villa et al1 report a study investigating whether the use of inline filters can reduce the incidence of phlebitis. For many, phlebitis is a minor aggravation, part of what happens every day in medicine, and is neither remarkable nor preventable. The mindset that the little things are not important is anathema to improving patient safety, and may contribute to preventable error or injury. Just as our chaos theorists have described how tiny events (the flap of a butterfly's wings in Brazil) can result in a large change (a tornado in Kansas), in medicine, we know that accumulation of minor events may degrade a team's ability to manage a major event.2 In 2001, Josie King, a delightful and beloved 18-month-old girl, was scalded on >60% of her body, and underwent weeks of burn treatment at Johns Hopkins Medical Center. During the course of her hospitalization, it became harder and harder to find peripheral veins, and eventually, a central venous catheter was placed. Tragically, just days before she was to go home, Josie developed an unrecognized bloodstream infection associated with that central line that eventually resulted in sepsis and death. What if it had been possible to prevent intravenous (IV)-related phlebitis in this little girl? What if a central line would not have been necessary at all? What if, by preventing phlebitis, we could have prevented her catheter infection and death? These questions haunt not only Josie's family, but also those who cared for her. Each year, nearly 200 million peripheral IV catheters are placed in patients in the United States.3 Many of the catheters need to be replaced due to infiltration, pain, swelling at the IV site, and phlebitis. Prevention of phlebitis is itself a worthwhile goal, and in Josie's case, could potentially have been life-saving. While the rate of bloodstream infections associated with peripheral IV catheters is just 0.5 per 1000 catheter days, the rate of phlebitis is more common (25%).3 Thus, with proper prevention of phlebitis, there could be longer catheter durability, fewer infiltrations, less patient discomfort, greater satisfaction, and perhaps a reduced need for central venous catheters. There are many factors that increase the risk of phlebitis, including insertion technique, infusion rates, and the type and character of infused drugs. One recognized, potential risk factor for phlebitis is the presence of microparticulates in IV infusions. In the late 1970s and early 1980s, there were ≥15 articles regarding use of inline filters to reduce the rate of phlebitis. These findings were not widely embraced. In 2001, Ball et al,4 returning to this "little" topic, described the volume and size of microparticles found in parenteral solutions, including plastic, glass, rubber, drug aggregates, and microbacterial particles. These authors reviewed multiple case reports in which these microparticles were felt to cause local inflammation, phlebitis, and arteritis, which, in some cases, led to death.4 The randomized controlled trial by Villa et al1 reports that inline filtration of all IV solutions reduced the incidence of phlebitis at 48 hours from 26.9% in the no filtration group to 2.2% in the filtration group. The study was performed at the Azienda Ospedaliero-Universitaria Careggi in Florence, Italy, where 268 surgical patients were randomized after the IV catheters were placed and before induction of anesthesia. The blinding of patients and the personnel who assessed the presence/severity of phlebitis was accomplished with a special dressing in both study groups. Catheters were evaluated every 12 hours up to 96 hours postoperatively. The reported results were promising, showing that at 96 hours, the incidence of phlebitis, based on the visual infusion phlebitis score of ≥2, was 58.2% in the nonfilter group and only 6% with the inline filter patients. This reduction in phlebitis at both 48 and 96 hours could alter practice given the simplicity of this solution to a common complication. Phlebitis leads to loss of that vein as an access point; being able to maintain IV access for longer durations could decrease need for additional IV cannulas as well as for more invasive access such as central venous catheters. Reducing the need for central venous catheters could inherently reduce the additional complications that accompany the risks associated with this procedure, such as infections, pneumothorax, nerve injury, and bleeding.5 Although randomized controlled studies of inline filtration support the reduction in phlebitis, and only a few studies refute the benefits, inline filtration has not become a standard of practice. Cost and convenience play a role in poor adoption, and few studies, including Villa et al,1 address the cost of these filters. A cost-benefit analysis would be useful in the future to guide hospital policies. There is also concern that having a filter in place would be detrimental during a resuscitative requirement. One very early study showed that having an inline filter led to a reduction in IV flow rates and incidents of filter blockage.6 The investigators acknowledged that while filters do slow infusion rates: the filters are quick and easy to remove in an emergency situation. In addition, Villa et al1 used ≥3 types of filters of different pore size: 1 for propofol and other lipid containing drugs; 1 for continuous narcotics; and 1 for all other drugs. Blood products would, in addition, require a separate filter. These requirements add to the complexity of IV catheter care in and out of the operating room. Moreover, in addition to cost and (in)convenience, a recent Cochrane analysis found that, although there was a reduction in phlebitis, there was no improvement in overall mortality in neonates when filtration was used.7 A somewhat older (2010) meta-analysis that included 11 trials for a total of 1633 peripheral catheters also found a reduction in phlebitis (odds ratio, 0.66; 95% confidence interval, 0.43–1.00), but concluded that the benefit was "uncertain" due to significant heterogeneity, with an I2 of 90.4.5 This heterogeneity was due to 4 studies that found a large treatment effect, while 7 other studies showed no treatment effect. Niël-Weise et al5 could not identify the source of the heterogeneity, so they were unable to endorse a benefit. The risk of publication bias is present in meta-analyses because journals are more likely to publish positive findings, and even a meta-analysis could be biased toward positive results. In the 2010 meta-analysis, the 4 publications with positive results had a very large effect size, which may have skewed the results. The current study by Villa et al1 indicates that a repeat meta-analysis may be due. Nonetheless, the dramatic results of this study provide cautious optimism. There were significant contributions to this study by grants from Orion Pharma, and Baxter also offered financial support for travel, hotel, and registration fees for conferences. While the study design seemed appropriate in randomization and blinding, conflicts of interest should always be considered when positive studies are funded by industry. Most studies of inline filtrations are older studies, during a time when conflict of interest was rarely reported. Despite some concerns with this study, it was well designed. Attempts to double blind the study seem both reasonable and effective. The investigators used special dressings to hide the presence or absence of the filter help to reduce bias. Although there are numerous definitions of phlebitis, the investigators used a validated visual infusion phlebitis score to attempt to objectify the diagnosis of phlebitis. Although numerous phlebitis scales exist and there is no consensus on measuring phlebitis,8 nonetheless, the authors used a prespecified and simplified scoring scale to reduce bias. It is clear that prolonging IV catheter life by reducing phlebitis can improve patient care. There are risk factors for phlebitis that cannot be controlled such as anatomic catheter site, which drugs must be given—especially parenteral nutrition—the volume infused, the duration of infusions, and frequency of bolus injections. Because these are uncontrollable risks, solutions for phlebitis include insertion technique, insertion location,9 maintenance infusions, and potentially, as reported by this study, use of inline filtration devices. This study shows promise and provides a simple solution that clinicians can control. If future studies confirm that inline filters reduce phlebitis and prolong IV catheter life, we can readily alter the "butterfly effect" on peripheral veins. DISCLOSURES Name: Chen Thay Chau, MD. Contribution: This author helped design, formulate, prepare, and construct the editorial, and attests to its review and approval of the final submission. Conflicts of Interest: None. Name: Richard C. Prielipp, MD, MBA. Contribution: This author helped design, formulate, prepare, and construct the editorial, and attests to its review and approval of the final submission. Conflicts of Interest: R. C. Prielipp is a member of the Speakers Bureau for Merck & Co, Kenilworth, NJ. Name: Joyce A. Wahr, MD, FAHA. Contribution: This author helped design, formulate, prepare, and construct the editorial, and attests to its review and approval of the final submission. Conflicts of Interest: None. This manuscript was handled by: Tong J. Gan, MD.