Blog

Todd Fraser wrote 07-24-2010 10:02:00 am
I do find this de-escalation thing difficult. Its relatively straight forward when you get a bug and can tailor your therapy, but what I find difficult is when you start and then have limited information to work with. For example, no cultures are taken in ED, the patient comes to your unit and 48 hours later is stable but not better. Cultures are negative - do you stop the ABs and reculture? How sick is too sick to do this?

The other situation I find myself doubting the strategy is in abdominal infections - you culture E. coli from a drain in a belly with peritonitis post colonic perforation. Okay, that's straightforward, but these infections are often polymicrobial. And anaerobic organisms can be difficult to culture. So do you continue broad therapy or not?

doug lynch wrote 07-24-2010 03:07:37 pm
We are working on an ICU antibiotic guidline in my Unit.

For the most part such a guideline is primarily aimed at reducing selection pressure for problematic resistance patterns. It is not an answer to individual prescribing conundrums.

Obviously each unit would only do this based on their own antibiogram.

Like so many things in ICU there will never be a one size fits all.

Do you use an anti-fungal routinely or was there a specific indication?

How does everyone feel about Gentamicin these days?

The brave might de-escalate this gentleman at two days perhaps to pip-tazo alone and then stop everything at 5-7 days and watch.

The source control point is of great importance. If all CVC, arterial, peripheral were removed then source control may have been achieved there.

And when removing a CVC do you guys routinely culture the tip, the line itself, one/two/three lumens?

It seems to me that unless you want to risk trying to avoid a CVC change then there is little benefit in culturing the CVC in any way. We just have to change all lines. Unless perhaps there is a CLABSI problem in the hospital and one wishes to quantify this.

If we are to culture the CVC there is an interesting paper I heard about on a podcast;

"How many lumens should be cultured in the conservative diagnosis of catheter-related bloodstream infections?"

Guembe M, Rodríguez-Créixems M, Sánchez-Carrillo C, Pérez-Parra A, Martín-Rabadán P, Bouza E.

Clin Infect Dis. 2010 Jun 15;50(12):1575-9.PMID:

Todd Fraser wrote 07-28-2010 06:11:16 pm
Thanks Doug,

Yes, the unit-based antibiogram is a very useful adjunct to prescibing practice. We've recently undertaken a similar thing in our unit.

The place of antifungals is a bit more perplexing. In this circumstance he had a number of risk factors for fungaemia (CVC, TPN, broad-spectrum antibiotics for >1 week, abdominal sepsis etc), so it seemed reasonable. Whether or not to start in patients with primary belly sepsis related to perforation is where I get a bit unsure.

As for Gent, I feel quite uncomfortable giving it these days. There was a study a few years back looking at balance difficulties after relatively short courses of Gent and the results were horrifying. I'll try to dig it out.

Recent literature seems to support taking cultures from all the lines (including the paper you cited, CID 2010; 50:1575-79). If there was no other obvious source of his infection I'd send the tip, but without clinical suspicion I think the results are less significant. Personally I would like to know my CRBSI rate, because meaningful things can be done to prevent it, so yes, I would chase the results.

Todd Fraser wrote 08-25-2010 10:30:06 pm
Another interesting debate came out of this patient's care. Following a laparostomy for raised intra-abdominal pressure and end organ failure, he developed a fever. His peritoneal swabs grew Candida parapsilosis. There was some discussion about whether or not this should be considered a coloniser given his abdomen was open. I felt it should be treated. What do you think?

Todd Fraser wrote 07-16-2010 10:24:23 am
So, not knowing much about this, I found an expert. Professor Jeff Lipman is the director of ICU at Royal Brisbane Hospital and has a keen interest in antibiotics in critically ill patients. Here's what he said :

"It is not clear if he was on antibiotics when he deteriorated or not. If so this complicates the growth of organisms (susceptible or not). Also prior antibiotic history is not provided - this always makes the choice of the next course of antibiotics more difficult and as well as possibly the duration thereof.

Please note that "source control" is vital in this case, even with a supposedly "normal" CT abdomen..... the adage "when in doubt blame the
surgery" has gotten me out of so much trouble, time and time again (never the blame the surgeon - makes collegiate work subsequently
difficult). Antibiotics are a very poor substitute for adequate source control!

Having said all that - if there is source control (and I am not sure about this in the case above), and pt is getting better with no
bacteria/fungi cultured (what happened to the E Coli and enterococcus?) I would give 5 days of the three agents and stop everything, watch the
patient and culture after 48 hours - note relatively arbitrary duration."

Jackson Bird wrote 10-21-2012 07:28:54 am
I would never consider the Candida as a coloniser in this circumstance. If its a swab from a "sterile site", it should be treated as true infection.

What happened?