Infection is a common occurrence in the ICU. When infections get bad, it almost always leads to something all of us dread - sepsis. This bodily reaction to infection is on of the core measures we look at in the hospital; it is a leading cause of mortality and can be devastating. The treatment of sepsis involves source control, and part of the sepsis 3 hour bundle involves putting people on appropriate antibiotics. But what antibiotics do you use?
This discussion will be directed toward someone who hasn’t been in the hospital and doesn’t know much about antibiotics or bacteria.
In the ICU, we worry about a couple subsets of bacteria due to the limits held by the antibiotics we give.
“above the diaphragm”
CAP - pneumococcus, H, influenza, pseudomonas, staph aureus
Peptococcus sp., peptostreptococcus sp., prevotella, veillonella, actinomyces
“below the diaphragm”
bacilli normal to the bowel; belong to the family Enterobacteriacae (“Enteric organisms”), gram negative rods. Generally a lot of anaerobic bacteria
UTI - E. Coli, proteus mirabilis, Klebsiella pneumonia
Intra-abdominal pathogens: urinary tract bacteria, plus interstitial anaerobes with includes Bacteroides fragilis, Clostridium Tetani, Clostridium difficile, bacteroides disastonis, bacteroides ovatus, bacteroides thetaiotamicron, fusobacterium
Pelvic infections: gonorrhea, chlamydia trachomatis, gardnerella
DRO (Drug resistance organisms) - think MRSA (methicillin resistant staph auerus), MSSA (methicillin sensitive staph auerus), CRO (carbapenem resistant organisms)
The two most common antibiotics you will see in the ICU for infection of unknown etiology are vancomycin and zosyn (piperacillin/tazobactam). This is because vancomycin covers gram positive bacteria, including your resistant organisms (think: MRSA) and zosyn has a broad range of gram negative and some pseudomonal coverage.
Vancomycin coverage: MRSA, staph epidermidis/coag neg staph, MSSA, enterococcus sp.)
Zosyn coverage: MSSA, enterococcus faecalis, streptococcus, anaerbobes (clostridium peptostreptococcus, bacteroides, fusobacterium), gram neg cocci, e. coli, klebsiella, proteus mirabilis, pseudomonas)
If you look below, you will see vanc/zosyn covers all of the organisms except for legionella (the bacteria that causes legionnaires disease/legionella PNA). This is why we like it so much. If you think someone has potential legionella (S&S of PNA + risk factors), you can cover them with vanc/zosyn/azithro.
Some specific trends you will see:
You will sometimes see patients be on Ancef (Cefazolin) for staph coverage peri-operatively
You will often see transplant patients on Bactrim for PCP coverage
Daptomycin can only be on board if ID is consulted
And some common medications and their effects:
Vancomycin can precipitate/worsen renal failure
Zosyn can cause thrombocytopenia
all about cultures
The most common cultures we get in the ICU -
Blood cultures (usually as part of sepsis bundle)
Urine cultures if deemed appropriate (a UTI can only be diagnosed with culture; not with a urinalysis, although a UA could lead you to suspect)
Respiratory cultures - sputum expectorated (patient spit out) or BAL (bronchioaleveolar lavage during bronchoscopy)
For each culture, you will see that the culture is either pending, growing “x” or no growth for “x” hours. Just because something doesn’t grow initially doesn’t mean it won’t pop up later.
But how do you know a culture is positive?
For blood cultures:
For urine cultures:
For respiratory cultures: you will get either quantitative or qualitative results
Quantitative, measured in colony forming units per mL/cc (e.g. “10,000 cfu/cc of E. Coli”)
If a BAL sample is at least 10^3 (>1,000), it is likely positive
If a mini-BAL sample is at least 10^4 (>10,000), it is likely positive
If a sputum sample is at least 10^5 (>100,000), it is likely positive
You must correlate with other clinical findings to treat it as positive (e.g. use a procal, S&S, sputum production, and other data)
Qualitative (e.g. “moderate growth of E. Coli”)
Usually you will treat no matter what; although lots of bacteria are in the lungs, they shouldn’t be there in an amount that shows up in large amounts on a culture
The team at the hospital will often read the culture as “normal respiratory flora” if normal bacteria are present
FCCS Ch. 11