So much of antibiotic selection has to do with local resistance patterns and it cannot be protocolized or generalized without that knowledge.
That said I believe there are some principles of antibiotic selection in the ICU that are worth discussing in general terms. A big shout out to Pharmacist Sokhak & Physician Mary for suggesting this topic. If you would like to suggest a topic for the show please do so here.
Today we will talk about:
- Source identification and control
- Knowing the antibiotic spectrum of activity
- Selecting empiric therapy against likely pathogens
- Renal dose adjustment of initial empiric therapy
Source identification and control
Source identification is generally taken care of by my docs and it is very rare that I find myself adding to the care of the patient in this area. The easy things to look for are:
– Blood cultures
– Urinalysis (>20 WBC/HPF)
– Chest X-Ray (infiltrate)
– Physical exam (belly, skin wound, etc)
Cold steel and the light of day cured many a disease. – William Conway
Your antibiotics aren’t going to do much if source control is not obtained. Know the patient management priority and if source control cannot be achieved due to a problem you can help fix – focus on that!
Blood cultures before antibiotics (when possible)
You can blow your nose and clean your glasses with the same tissue, but you better do it in the right order.
Advocate for blood cultures prior to antibiotics, but don’t withhold the antibiotics if it looks like blood cultures will not be readily obtained. Surviving Sepsis guidelines recommend no more than a 45 minute delay in antibiotics for the purpose of obtaining blood cultures first.
The rare case where you would run into trouble here is giving antibiotics prior to cultures in a patient that turns out to have endocarditis. If their cultures come back negative because of this they may need 6 weeks of broad spectrum therapy – yikes!
You’ll never forget what to look for as far as source identification after watching this video:
Knowing the antibiotic spectrum of activity
I’ve found it is much easier to learn the rules and then learn the exceptions to the rules when it comes to knowing what drugs cover what bugs.
This section will not be comprehensive but will focus on classes of antibiotics likely to be used as empiric therapy. I encourage you to get information on dosing, mechanism of action, etc… elsewhere. I suggest John’s Hopkins Antibiotic Guide (unless you need pediatric dosing in which case I suggest the Sanford Guide).
MRSA is covered by vancomycin, linezolid, daptomycin and ceftaroline.
- Daptomycin is inactivated by lung surfactant – don’t use in pneumonia
- Linezolid is bacteriostatic – don’t use in bloodstream infections
- Ceftaroline is vancomycin and ceftriaxone rolled into one loveable fuzzball
Piperacillin-tazobactam and the carbapenems ertapenem, meropenem, doripenem, and imipenem cover gram positives, gram negatives, and anaerobes.
- Ertapenem doesn’t cover pseudomonas auruginosa (hereafter refered to as pseudo)
- Piperacillin-tazobactam doesn’t cover ESBL well enough to be used in severe infections (it probably works for UTI thanks to @ASP_PharmD for pointing this out)
Cephalosporins for empiric use in the ICU are cefepime, ceftriaxone, ceftazidime, and ceftaroline. They cover gram positives and gram negatives.
- Only cefepime and ceftazidime for pseudo
- None cover enterococcus (remember this shows up on the gram stain as “gram positive cocci resembling strep!”)
Quinolones ciprofloxacin, levofloxacin and moxifloxacin cover gram positives and gram negatives (double check your local resistance patterns!!!).
- Ciprofloxacin is weak against strep pneumo, but causes the least qtc prolongation and probably doesn’t cause torsades (for a great reference on QTc interactions I highly recommend the book Top 100 Drug Interactions by Hansten & Horn)
- Moxifloxacin is best against anaerobes but has no pseudo coverage
- Moxifloxacin is hepatically metabolized – this is a double edged sword as it has no dose adjustment in renal failure but can’t be used for a UTI
Selecting empiric therapy against likely pathogens
You need a ‘cillin to do the killin’.
Whenever possible I am looking for a strong, bacteriocidal antibiotic to give first as empiric therapy. Let’s kill some bugs!
Community infection – ceftriaxone plus azithromycin or levofloxacin/moxifloxacin
Hospital infection – vancomycin plus piperacillin-tazobactam or cefepime
Aspiration – add anaerobic coverage
Community infections – the likely pathogens differ by age group – vancomycin, ceftriaxone, + ampicillin if over 50 years of age
Hospital or immune compromise – vancomycin, cefepime, and ampicillin
This is a surgical emergency! Don’t do anything that will delay the patient’s transportation to the OR!
Pathogens include group A strep, anaerobes, MRSA, and the occasional gram negative – use linezolid + piperacillin/tazobactam or vancomcyin + piperacillin/tazobactam + clindamycin
Clindamycin or linezolid should be part of every regimen until group A strep is ruled out but their purpose is to shut down toxin production. They cannot be used alone due to their bacteriostatic nature.
You’ll need to hit gram negatives as well as anaerobes – use piperacillin/tazobactam or meropenem/imipenem/doripenem
A community acquired infection shouldn’t need empiric antifungal therapy. Nosocomial infections may – if you have a critically ill patient with candida isolated you should use micafungin until you know that fluconazole will be OK.
Vancomycin + piperacillin/tazobactam or cefepime or meropenem – the sicker the patient the broader the coverage
Your local resistance patterns dictate this. Surviving sepsis guidelines do recommend (level 2B) empiric double coverage against pseudo in neutropenic patients for 3-5 days unless treatment can be narrowed sooner.
Other bugs once identified also have a 2B recommendation for double coverage (pseudo and strep pneumo). Your antibiogram should determine whether the 2nd agent is a quinolone or aminoglycoside.
Renal dose adjustment of empiric antibiotic therapy
There are two reasons to adjust a medication for renal insufficiency. To prevent toxicity or to save money.
There will be plenty of opportunity to reduce drug costs if the patient survives. If I am using a relatively non-toxic antibiotic (think anything with a beta-lactam ring) I will dose as per the patient’s baseline renal function for the first 24 hours and then re-assess the dose.
Remember the serum creatinine lags actual improvement in kidney function and Cockcroft-Gault studied mostly 50 & 60 year olds who all had stable renal function.
To help remember spectrum of activity, I’ve created a visual antibiotic guide. You can download inside the free Pharmacy Nation Community.
If you like this post, check out my book – A Pharmacist’s Guide to Inpatient Medical Emergencies: How to respond to code blue, rapid response calls, and other medical emergencies.