Infectious Disease Compendium

Catheter Infections


IDSA guidelines if you have insomnia. Cancer guidelines for a coma.

Draw simultaneous cultures from the catheter and peripherally, if the set from the catheter is positive 120 or more minutes before the peripheral set, the catheter is probably infected and the source of the bacteremia. If they are positive at the same time, then the catheter, while it may be infected, is probably not the source. This assumes that your have continuous monitoring of blood cultures.

And obtain a culture from each of the ports (PubMed) or you will under diagnose the infection.

You can also pull the catheter and do a culture of the tip; if > 25 cfu by the Maki method, then the catheter was infected. Note the was. The catheter is now gone.

Pay attention here: if the catheter tip is positive for Staphylococcus aureus BUT the blood cultures are negative, 24% of patients will have a subsequent S. aureus bacteremia; if promptly treated the risk of bacteremia is decreased 84% (PubMed)(PubMed).

Also, if the line blood cultures are positive, but the peripheral blood cultures a negative, the patient has an 8% (more for Candida than bacteria) risk for future bacteremia (PubMed).

If the time to positivity for Candida is < 30 hours, it is more likely a catheter infection as the source of the Candidemia (PubMed), you remember that if S. aureus has a time to positivity of < 24 hours, it is more likely and endovascular infection. However, a catheter tip for Candida is only rarely associated with fungemia: about 2.7 % of cases (PubMed).

Epidemiologic Risks

I think having a catheter in your vein just may be a risk. Catheter infections have many types: hub infections, tunnel infections, insertion site infection, lumen infection and intravascular infection are all possible and can have different manifestations.

In dialysis patients, finding coagulase negative Staphylococcus endoluminal catheter colonization increases risk of blood stream infection in the next month (PubMed).

There are 80,000 catheter infections a year and 20,000 deaths and PREVENTION IS SIMPLE (PubMed): hand washing, using full-barrier precautions during the insertion of central venous catheters, cleaning the skin with chlorhexidine, avoiding the femoral site if possible, and removing unnecessary catheters. Doesn't seem all that hard, does it? And it really really works. And if you combine a prevention bundle with a check list (PubMed) so you make sure you remember to do everything in the bundle your central lines infections should almost vanish.

Curiously, in a study that may or may not be validated, 325 mg a day of aspirin was protective for S. aureus catheter infections in dialysis patients (PubMed).

Summer time is catheter infection time in the outpatient setting, probably because most people do not have air conditioning and get all sweaty around their line (PubMed).


Coagulase negative staphylococcus lead the list, but any organism can cause an intravenous catheter infection. If the BC are positive in <16 hours, it is more likely real; if BC are positive > 20 hours it is more likely a contaminant (PubMed).

In patients with hematologic malignancies, gram negative rods may predominate (PubMed).

Transplant infections can get Mycobacterium mucogenicum. Atypical mycobacteria can occur with long term (> 6 months) access including Mycobacterium mucogenicum, Mycobacterium fortuitum, Mycobacterium neoaurum and Mycobacterium septicum; all of the patients had a long-term intravascular catheter (mean duration, 6.5 months (PubMed)).

If cultured anaerobically, Propionibacterium acnes may account for 15% of catheter infections (PubMed).

B. cerus can cause line related infections, in Japan it is a summer disease (PubMed).

Molds can infect the tunnel on cancer patients.

Empiric Therapy

Vancomycin plus a third generation cephalosporins are not unreasonable.

Specific Therapy

Pull the line. Almost always. Maybe a coagulase negative staphylococcus and probably a streptococcus can be salvaged. Everything else should have the line come out. Do not try to salvage lines infected with S. aureus, gram negative rods (see next paragraph), Mycobacterium or Candida. You will fail and the patient can have any number of complications.

As a rule, the longer you wait to pull the catheter, the greater the mortality (PubMed).

For dialysis catheters "suggest that tunneled hemodialysis catheter­ related bacteremia should be treated with either guidewire exchange or antibiotic lock solution", although for S. aureus there were better outcomes for an exchange (PubMed). My brilliant idea is of you are going to do a guidewire exchange, do it through either an alcohol or antibiotic lock.

Although one study of uncomplicated gram negative line infections had a 95% cure rate with a combination of iv antibiotics and an antibiotic lock (PubMed). In my experience, an alcohol lock with iv antibiotics works as well.

If you are going to try and salvage a line infected with Candida, there is in vitro data to suggest that caspofungin infusion and antibiotic lock MAY work. Caspofungin and its ilk kill yeast in biofilms.

Duration of therapy after line pulled? For S. aureus a minimum of 2 weeks IV. Never ever treat S. aureus bacteremia with po unless you are a) stupid and b) want to get sued. The IDSA suggests 5-7 days for coagulase negative staphylococcus, and 7 -14 days for enterococcus and gram negative rods (PubMed). The duration for coagulase negative staphylococcus and enterococcus are based on expert opinion not data. Opinions are like ass holes. Everyone has one, and everyone's stinks. I think it is way over treating in normalish hosts, but whatcha gonna do?

If the infection is due to coagulase negative staphylococci or a streptococci the line may be salvaged with 2 weeks of IV therapy. Several studies have suggested benefit in adding N-acetylcysteine to the antibiotic lock to break down the biofilms (PubMed). There is one study to suggest salvage is not as easy to accomplish as suspected if you use strict criteria for line infection (PubMed).

There is at least a 20-30% chance that S. aureus from a line will go elsewhere, such as a valve. A TTE will miss 20% of vegetations that are found on subsequent TEE.

Pay attention here: if the catheter tip is positive for Staphylococcus aureus BUT the blood cultures are negative, 24% of patients will have a subsequent S. aureus bacteremia; if promptly treated the risk of bacteremia is decreased 84% (PubMed)(PubMed). This is not true for other organisms (PubMed). However, a catheter tip for Candida is only rarely associated with fungemia about 2.7 % of cases (PubMed).

Linezolid is not inferior to vancomycin (PubMed), but is one hell of a lot more expensive.

The general rule of thumb with infections is the sooner the correct therapy and the sooner the source is controlled (in this case, pulling the catheter) the better the patient will do. What passes for integrity in my case leads me to note that pulling the line for Candida catheter infections early didn't make a difference in outcome (PubMed); I would still do it.


Even after you pull the central line the residual fibrin sheath can be left behind and be a source for continued infection.  Need a TEE to diagnose it (PubMed).

As a rule, it is a waste of time to pull central catheters for just a fever in the absence of positive blood cultures. Document infection first (PubMed). Occasionally in the long term neutropenic, if you are suspecting a difficult to grow mould, or in a patient with a prosthetic heart valve where you cannot risk the valve being seeded, you may pull a line because of fever. But usually it is a waste of time.

For peripheral iv's, routine changes had no fewer complications than waiting for clinical indications (Pubmed). And do not try to change potentially infected lines over a wire. It is against the guidelines, and you would not want to go against the guidelines, now would you.

If you have a patient with recurrent line infections had has to have the line, try an alcohol lock: 25% ETOH placed in line (silicone only) for 30 minutes 3x a week, then either withdrawn or pushed through. A meta-analysis supports the practice (PubMed). Some catheters cannot tolerate ETOH (polyurethane); not approved by AA.

Change an infected catheter over a wire? The data is variable but suggests that it is no worse than a new line. Still, it gives me the Heebie Jeebies to put a sterile wire into an infected catheter and expect the whole system not to get infected. If it has to be done, my non evidence based suggestion (anyone want to do a study?) is to fill the catheter up with 25% alcohol before sticking a wire in it.

Antibiotic locks can both treat (PubMed) and prevent (PubMed) infections, and do not (allegedly) breed resistance (PubMed); not all studies agree (PubMed) a meta analysis here (PubMed). I prefer alcohol locks, which does not breed resistance and the bacteria die happy (PubMed). Just fill the catheter up with 25% alcohol for 30 min once a week, then either push it through or draw it out. Not all catheters can be alcohol locked. One in vitro study suggested 30% alcohol for at least 4 hours (PubMed).

There is ZERO data to support dental prophylaxis to prevent infections of catheters (PubMed).

I had one patient with a Groshong who came in for a catheter flush every month. She would get home, have a rigor and a fever then be fine. This went on for about half a year when she finally mentioned it to the nurse, who check blood cultures through the line and it grew Enterobacter. The infection was in the lumen of the line and she only got sick when they pushed the bacteria into the vascular tree with the flush.


All central lines should be places with full sterile barriers (gown, mask, glove, drapes, the whole shootin' match). If you are not aware of these recommendations, based on controlled studies, than you are bigger fool than you look and I recommend getting up to speed on proper technique before you take care on any more patients.

Curious Cases

Relevant links to my Medscape blog

What's My Line? Infection

Playing from the tips

Bread and Butter

Dirty Laundry

Make My Life Easier

From the Tips

Pull or Wire

Stripped Off


Last Update: 08/12/18.