- Meet Consumers/Patients
|Newsletters: Catheter Related Infections|
Catheter Related Infections
Barry M. Farr, MD, University of Virginia Health System
Dr. Farr presented a tremendous amount of data from the latest research on catheter related blood stream infections at the 2003 Oley Conference. An overview of his talk is published below; the talk is available in its entirety from the Oley DVD/Video Library. A copy of the detailed “Guidelines for the Management of Intravascular Catheter-Related Infections” co-authored by Dr. Farr is also available by sending a stamped, self-addressed envelope to the Oley Foundation. Many thanks to Dr. Farr for sharing his time and expertise, and to Robin Lang for transcribing the presentation.
Different catheters require different management for catheter-related infections (CRBSI). First it is important to note which type of CRBSI is more common with which type of device. CRBSI in short-term, non-tunneled catheters are often related to extraluminal (outside the catheter) colonization of the catheter which originates from the skin, most likely because the lack of internal attachment allows the catheter to move in and out a little; they are less commonly infected by intraluminal colonization of the hub and catheter lumen. In contrast, the most common route of infection for long-term, tunneled catheters and implanted ports, is intraluminal (within the catheter). With intraluminal infection, the exit site and tunnel (skin area overlying the catheter) appear to be normal, but fever/chills are present and symptoms may be worse when infusing.
The most widely used lab technique for confirming the diagnosis of CRBSI is a semi-quantitative catheter segment culture method. A 5 cm catheter segment (usually the tip) is rolled across an agar plate 4 times, then the colony forming units (bacteria/fungi) are counted over a 24- to 72-hour incubation period. Concern has been raised that the coating on anti-microbial coated catheters could sometimes lead to false negative results when a culture is taken. Quantitative catheter segment culture methods are somewhat more accurate than the semi-quantitative method, but are also somewhat more difficult and expensive to perform. They involve a technique such as “sonification” of the catheter segment in some culture broth. This removes microbes from the catheter and allows the broth to be cultured quantitatively. The main drawback with both of these methods is that they require removal of the catheter, which may be salvageable or not even infected.
Since access can be an issue for long-term catheter users, an alternative method has been developed that uses quantitative cultures of blood, one set drawn through the CVC, and another percutaneously from a peripheral vein. A colony count from the catheter sample that is at least 5 to 10 times greater than the peripheral sample would indicate CRBSI. In tunneled catheters, for which the method has seemed most accurate, a quantitative culture of blood from the catheter that yields at least 100 cfu/mL may indicate a CRBSI without comparing it to a peripheral blood sample.
A new method under investigation is called “differential time to positivity.” This technique also requires the culturing of blood samples drawn from the catheter and a peripheral vein. CRBSI is indicated if the catheter sample becomes positive more than two hours before the peripheral sample. In several studies, this test was found to be as accurate as the quantitative blood culture method. It may be somewhat quicker to use than quantitative blood cultures and more hospitals will have this method available than the quantitative blood culture methodologies.
Another approach being developed uses an endoluminal brush (FAS Medical) to clear the inside of the catheter lumen and procure a sample of the biofilm lining the catheter. (See article describing the endoluminal brush). The sample is taken to the lab for culture. If the infecting agent can be identified, there is a better chance of clearing the infection and saving the catheter. The drawback to this technique is that it requires training for the clinician, and isn’t widely available yet.
When there is concern about endocarditis, a trans-esophageal echocardiogram (TEE) should be done to make sure this complication isn’t present in addition to the CRBSI.
Common Culprits of CRBSI
The four most common microbes to cause CRBSI are:
1. Coagulase negative staphylococci, a bacteria present on the skin of all human beings. Such infections have a very low patient mortality rate (0.7%) and 90% of catheters with this type of infection can be cleared.
2. Staphylococci aureus, a bacteria associated with a significantly higher patient mortality rate (8.2%). Methicillin-resistant Staphylococcus aureus (MRSA) is associated with an even higher mortality rate, often because the therapy is less effective against the microbe. When MRSA becomes vancomycin resistant as well, this has resulted in a very high patient mortality, again because the therapy is less effective against these microbes. There is a high probability that a catheter infected with this microbe will be pulled.
3. Gram negative bacilli mostly come from the gastrointestinal tract; others come from aquatic sources (e.g., tap water). A physician might try to clear this type of infection before determining whether to pull the catheter.
4.Candida species, a fungus which like Staphylococci aureus is associated with a high patient mortality rate. It is part of the normal flora of the gastrointestinal tract. With these types of organisms, the catheter is almost always removed.
Most short-term, non-tunneled catheters are considered temporary lines and are usually pulled when CRBSI is suspected. For tunneled catheters and ports, the decision to pull the catheter is more difficult. The decision depends upon the patient’s need for the particular catheter, the severity of the CRBSI, the type of microbe involved, and possible complications (i.e. septic thrombophlebitis, endocarditis and/or metastic seeding).
When CRBSI is suspected, cultures are drawn and then IV antibiotics are typically started right away. Vancomycin is frequently recommended because most coagulase negative staphylococci will require this therapy. Once culture and antibiotic susceptibility results return, a decision is made about whether to pull the line and what type of antibiotics to use. Nafcillin or oxacillin should be used for methicillin-susceptible staphylococci. Amphotericin B or fluconazole are the usual choices for candidemia.
How long the patient must remain on antibiotic therapy depends on the microbe and the anatomic extent of the infection. A simple CRBSI would typically receive a 14 day course, whereas a patient with evidence of septic thrombosis, endocarditis, osteomyelitis or metastic seeding, may require 4 to 6 weeks of treatment. Streptokinase has been used in combination with antimicrobial therapy, but its use has not been shown to be beneficial as an adjunctive therapy to antibiotic treatment among patients with CRBSI whose catheter is not removed. TPA has also been used.
When the catheter is left in place and an intraluminal CRBSI is suspected, two weeks of antibiotic lock therapy is recommended in conjunction with the systemic antibiotic treatment to prevent a recurrent infection. Remember, most CRBSIs with tunneled catheters involve the spread of infection from the catheter hub to the catheter lumen. A much higher concentration (100 to 1000 times greater) of antibiotics is necessary to kill bacteria in a biofilm than in solution/blood. Both facts point to the need for an antibiotic lock: e.g. 1 mg/ml concentrations of vancomycin should be instilled in the infected catheter for hours or days when the catheter is not in use. Using this technique has resulted in significantly more catheters being saved, except with certain pathogens like fungi.
A number of different factors described below have been shown to influence the risk of CRBSI.
Where the catheter is placed has a lot to do with the risk of infection. Catheters placed in the subclavian have a lower risk of CRBSI than those placed in the internal jugular or femoral veins. Femoral veins are also more prone to thrombosis (clotting).
It is recommended that a maximum level of sterile technique be used when catheters are placed. Clinicians should use large drapes, dress as for a surgical procedure (i.e., cap, mask and a sterile gown and gloves), after cleansing hands with antiseptic soap or an alcohol solution. The patient’s skin should be cleansed with a chlorhexidine solution. These precautions have all been shown to lower the risk of CRBSI. CDC recommends this approach.
When first introduced, transparent dressings were very occlusive. They did not allow water in; however, they did not allow moisture out either, which left a breeding ground for microbes. In studies, transparent dressings have been linked to more microbial colonization at the exit site, which lead to CRBSI. There is still controversy as to which type of dressing is better: transparent dressings, or a gauze and tape dressing. Current transparent dressings are more breathable, and shouldn’t be relied upon as a moisture barrier when the patient showers. (Editor’s note: a limited number of sample moisture barrier dressings were recently donated to the Oley Foundation and are available on a first come, first serve basis by sending a self-addressed stamped 9” x 12” envelope to The Oley Foundation, 43 New Scotland Ave., MC-28, Albany Medical Center, Albany, NY 12208)
Long-dwelling catheters (e.g., tunneled catheters) have a lower risk of becoming infected from microbes outside the catheter. One infection per two catheter years is the average infection rate for an adult, long term TPN consumer. Catheters that are no longer needed should be removed.
Studies have shown that catheters reserved only for TPN, not multiple uses, are less likely to become infected.
In hospitals, the administration set should be changed every 3 days for TPN, and within 24 hours when lipids are used.
Several studies demonstrate that having only IV specialists handle the catheter is safer than a whole myriad of health care workers. Don’t hesitate to remind staff if they are not wearing gloves and/or practicing proper technique. IV specialists have better training regarding catheters and catheter care.
To protect health care workers, a needleless system is being used throughout the United States; however, there have been multiple reports of outbreaks of infections in patients due to these devises. Studies showed that end caps were not changed as frequently as recommended (2 times per week). When used properly, the needleless systems have usually had CRBSI rates comparable to those of the old needle systems.
Downsizing of hospitals has put patients at increased risk for infection. Smaller infection control teams have often meant less stringent surveillance for infections. Meanwhile there have been too few nurses, especially registered nurses, in many facilities. As a result, CRBSI rates have gone up in some facilities, as well as health care costs. “An ounce of prevention is worth a pound of cure;” investing in infection control teams saves money.
Using chlorhexidine to prep the catheter site (before insertion and every few days during the first weeks that a catheter is in place) has been shown to work better than alcohol or povidone iodine for reducing the incidence of CRBSI. However, antibiotic ointment, i.e. neosporin, bacitracin or PNB ointment, should not be regularly used at the catheter site because it increases the risk of Candida colonization 5-fold.
Antiseptic (silver sulfadiazine-chlorhexidine) coated catheters have been shown to reduce the risk of infection. Antibiotic-coated catheters (minocycline-rifampin) also work, but concern has been raised that they will encourage the growth of antibiotic resistant microbes.
Vancomycin locks, have been used to prevent CRBSI, but the CDC recommends against this practice due to the risk of developing antibiotic resistance, particularly vancomycin-resistant strains of microbes.
In summary we need to:
Questions & Answers
Q. “When blood cultures are taken from the catheter, should the first vial of blood drawn off be thrown away?”
Dr. Farr: “If you are looking for an intraluminal infection, some believe that the first vial should be saved, and a second one taken as well. Typically the first vial of blood drawn off is discarded because it may contain the solution that was last being infused through the catheter and that material could interfere with the accuracy of the blood culture (especially if it included an antimicrobial). If it doesn’t contain an antimicrobial, however, and represents fluid that had been sitting in the catheter for hours, it may contain a higher concentration of microbes.”
Q. “When do I need to see my physician or go to the hospital?”
Dr. Farr: “If you have symptoms, such as fever or chills (rigors) during infusion, it is important to be seen at that time. Positive cultures are more likely to be found when you are symptomatic.”