April 7th, don't be defenceless against sepsis!

Starting the rotation for incoming residents on Tuesday, we discussed a patient who had come in febrile (38.6), HR 1005, and bp 90/60.

I stressed the importance of looking for fever in ALL your patients as it may tip our differential in one way or the other. Whether you are calling nursing homes, looking at form browser for true triage vital signs, or vital signs in emerge to see if they ever went above 38. 

*Note, patient saying they had sweats, chills and fever is still worth documenting, but that would be reported as "subjective" fever.

Fever of Unknown origin was previously talked about, and the differential is long, most large studies would usually list infection/connective tissue disease/and malignancy as the top three. 

Some simple criteria

SIRS: HR >90, RR >20 (or PCO <32), WBC >12,000 or <4000 or >10% immature neturophils, Temp >38 or <36.

All of the above need to be taken in context! A patient with CLL may have a "white count"of 40,000.  A patient with COPD may have a baseline CO2 of 60 and coming in at 52 is below their norm.  A patient with afib may sit at a heart rate of 100, and may be coming in at 120 with rapid afib. A patient without a fever maybe taking tylenol around the clock at home for arthritis! 

We do not include SIRS criteria in blood pressure, probably just because it wasn't studied particularly for this, but make sure we use it! Also important to consider a patient's baseline bp. Maybe there normal is 160/100 and coming in at 120 is an early sign of sepsis. 

So what is SEPSIS? Easy - SIRS plus a source of infection - thinking head to toe, meningitis, mouth infections, lungs, abdo (gallbladder, liver, colon, small/large intestine), heart, urine, bones, skin, LINES! Do not write your admission orders until you have at least a suspected source, as one of the most important things we MUST do is achieve source control (taking out infected lines, ERCP for a stone blocking the bile ducts, etc)

Sever Sepsis: involves end organ damage DUE to sepsis, common things we look for is lactic acid (tissue hypoperfusion), increased creatinine, trop elevation. Uptodate has an extensive list, you have to be careful not getting overwhelmed with this and trying to assess for older baseline values.

Septic shock refers to hypotension SBP <60 OR < 80 if they have baseline hypertension and unable to keep MAP >65 with adequate fluid resuscitation.

As you can imagine, mortality increases as you move down the latter form sepsis to septic shock.

The surviving sepsis campaign is here http://www.survivingsepsis.org/Guidelines/Pages/default.aspx

This largely came from the Rivers trial in NEJM in 2001. They randomized patients to goal directed therapy

(CVP 8-12)

urine output >0.5 mL/kg/hr,

MAP >65

ScvO2 >70%, maintained with pressors or blood cell transfusion.

 Subsequent trials (PROCESS and ARISE) have questions central venous pressure/ ScvO2, as well as meta-analysis published have caused some ICU physicians to render it useless. Keep in mind that two large bore IVs can resuscitate much fast than a central line, but we generally will not want to give pressors through a peripheral IV.

ScvO2 obtained from blood from the right side of the heart has also been questions as a target for goal directed therapy.

For R2s and above, this is a meta-anlaysis and review around CVP please read http://www.ncbi.nlm.nih.gov/pubmed/18628220

Author Paul Marik actually trained in London, he has published extensively as an intensivist.

Our patient presented to us had a PICC line that we believe was the source of infection, the reason why gram positives have taken over as the most common cause of sepsis. Remember, they don’t have endotoxins like gram negatives do, and might not present as septic shock.

So when you suspect infection, what do you do? Broad spectrum initially, vanco would be standard of care if suspected gram positives, and piptazo is reasonable if suspecting gram negatives. 

Looking through power chart for previous bugs grown in blood, urine, feces, or for MRSA colonization will help you decide what infections they can have!

Two BIG things not to miss!

1. History of MRSA colonization - will need antibiotics with MRSA coverage! 

2. History of gram negative ESBLs - this stands for extended spectrum beta lactamase inhibitor, and are resistant gram negatives, carbapenems are the drug of choice and you would start this empirically as pipatazo will NOT cover these.

Some other points to mention: sugar control, Hb levles are recommended in the surviving sepsis guidelines, largely answers come from the TRICC trial (increased mortality with giving more blood, and the NICE sugar trial - don't go to low!)