Sepsis Questions

Sounds like homework...

I wish it were just homework. New grad in the ICU with little exposure to sepsis.

1) A pt with sepsis has a 101 temp. Which is more appropriate at treating this: a cooling blanket or acetaminophen? Both are avail for fever.

2) Corticosteroids with vasopressin or Levo/Dopamine if needed longer than 6 hr. Rationale?

3) Pt has a decrease in SBP from 120s to 86 in 30 min. A/Ox4 but sleepy. Pt received 1L NS bolus in ER for SBP 80s 2 hours ago, has hx of CHF, in for pneumonia and poss sepsis, on requiring BiPap. Sepsis bundle has NOT been ordered for pt. After waking pt up SBP 110s, 1 hr later SBP 82. Pt husband reports "BP runs low". Okay to request Levo just in case or is that overthinking it too much?

1. Wouldn't treat a 101 temp unless patient is uncomfortable or for some reason couldn't tolerate the increased metabolic load. Cooling blankets are very uncomfortable and the shivering can actually make things worse. If I felt I had to treat that fever I would use Tylenol.

2. Neither. Vasopressin at .04 can be used for sepsis but steroids rarely have a place. Norepinephrine is a great drug for sepsis but I am not a fan of dopamine.

3. If truly septic/SIRS she will need more than 1 liter of fluid. Needs to be volume resuscitated before giving pressors.

Others may disagree, these are just my thoughts.

101 is not that significant of a fever. I wouldn't bring out the cooling blanket for that.

1) Killing off a fever isn't always a good thing. One of our fellows did a paper on this a while back. If they start showing some decompensation from a fever (tachycardia, excessive sweating leading to fluid deficits) then it might be a good idea to treat it. Acetaminophen first, cooling blanket for a fever unresponsive to conventional pharmacological measures.

Don't forget to check your LFT's, shock liver = no Tylenol. Elevated transaminases alone does not contraindicate Tylenol, but if your liver is currently doing squat for your body, might want to hold off. I'd like to see some research proving the validity of this statement, but it's what we roll with on our unit.

2) Steroids - recent sepsis guidelines state they have no place unless you've absolutely tanked up the patient with fluids and they are unresponsive to pressors. Levo is great for septic shock as it provides both beta and alpha agonism, and in septic shock (and typically gram negative) you have some cardiac dysfunction due to the presence of MDF along with profound peripheral vasodilation.

3) BP is just a number. I've seen people sleep with MAP's in the 40's, wake 'em up, they're fine. Do you have a rationale for being concerned? Look at urine output, lactate, mentation, etc. If the family says "they run low" but their lactate is 5 and they've made 30 mL of urine in the last 3 hours, might wanna treat it. Again, tank them up with fluids first, and if they don't respond or you hear crackles from 3 rooms down, then you may consider going to pressors. CHF and sepsis is a fun mix, but they absolutely must be given fluids first in any situation. Can't clamp down on empty arteries and expect them to work well.

@detroitdano: BEST answer yet! Thank you so much!

First thing I do for a fever is remove all the blankets the family has placed on the bed. I turn down the A/C, I remove the 10 pillows from all over the bed, and turn on a fan. If pts are wrapped up like a burrito, the tylenol only works for a little, then it goes back up. A significant percentage of low grade "fevers" (I know this post is about sepsis, but just to make a point), can be caused by the environment. Get in bed with a shirt, socks, and pants (because that's basically what the SCD stocking are) place pillows under you legs, and put pillows under your arms so they bunch up against both sides of your body...and then put a blanket over you. Imagine being like that all day. It's like an oven. You can keep body temps down a full degree or two by just removing some of pillows and blankets a couple times a shift. That's why bed baths can cool down the body. Along with the water that's evaporating from the body, the removal of the pillows, blankets, socks, and scd's expose the skin to the air. Many pts only have the head exposed and no wonder everyone has a temp of 99.5.

If there is a problem with temperature regulation (that can't be corrected by an environment change), I'll go with tylenol. I stay away from cooling blankets unless the body is at risk for temps over 103. Cooling blankets are horrible for skin care. If you want to speed up the process of decubs, use a cooling blanket. Or, if a pt has head trauma and I know that tylenol will do nothing for a fever, then I will sparingly use the cooling blanket. I've never used a cooling blanket on a septic pt unless they have malignant hyperthermia. A removal of pillows and use of fan does wonders.

2) Never use vasopressin for sepsis. I known that bundles have that order, but in my opinion, vasopressin should be removed from the bundle. The only time vasopressin should be used is when all other vasoactive gtts are maxed out or the pt is in DI. It drives me absolutely insane when I get a septic floor pt who was automatically put on vasopressin. "Well, the pt's urine output basically stopped, so I think he's now in renal failure." No, it stopped b/c you put him on vasopressin. That's the side effect of a continuous gtt of antidiuretic hormone. I immediately wean off the vaso and put up Levo, and like magic, there's output again. I think people like vaso because it's really not much of a titrated gtt. They put it on at .04 and move on. It's easiest.

Hydrocortisone is used when the pt is hydrated and the pressors are starting to get maxed out (but this is debated). Although, I think it sometimes works. Levo and or even Neo is best for sepsis. I don't like dopamine b/c of the tachycardia. You should know the true heart rate.

When someone has labile b/p I think that is an indicator that the pt is not well hydrated. When someone's bp is low, the body has ways to temporarily increase it. Angiotension, ADH release, epinephrine release, etc. The body reacts to bring it up, but there isn't a cushion of volume. The up and down motion of the b/p is the body doing it's best can to raise it. So, you push fluids. A pt in true septic shock can pool up to 50% of volume in the capillary beds. Some pts can require 2-5 liters to just keep things moving. This is where a CVP monitor comes in handy. You can give someone 2 liters of fluid and still have a cvp of 2. Throw a non-valved PICC in and that can somewhat give you an idea of the reading. A CVL is best, but a PICC can do the job.

I don't listen to people when they say their blood pressure "runs low." Sure, when you are at home and healthy or you are in the hospital for a hernia repair. When you have someone that is septic, it's a different ballgame. I don't care how things are normally, but I'm not happy unless your systolic is greater than 95. Sometimes, I shoot for greater than 100. If I have a little of lady with CHF, it's good to have levo on the MAR. Give her what she can handle, then use levo b/c you will be getting an order for lasix after putting her on the vent. Then there's no right answer b/c you'll be putting her on levo anyway after the lasix drops the pressure.

What else did you try for the fever? Take the blankets off, make the room cool, etc. Tylenol would be my next choice. And no patient that is awake and alert is going to take the cooling blanket. Haha. It's not always appropriate to break a temp if the patient is tolerating it.

The literature is pretty clear that levophed is the best pressor for a patient needing pressor support on the sepsis spectrum. Vaso is usually reserved for sepsis patients but only when they are in severe septic shock that is refractory to max dose levo/neo.

Take a look at the algorithm for early goal directed therapy in sepsis. It recommends treating the CVP first with fluids to a CVP of 8-12. Then treating the MAP with vasoactives to a MAP of >65. Then you treat the ScvO2. Ionotropic agents like dopamine are usually reserved for low CO or EF (ex: if a patient's EF is

But of course you don't always get the CVP or an A-line right away, etc. So you have to know how to look at lactate and UOP and the entire clinical picture.

One liter? Without some indication other than BP what their volume status is you are flying blind but if they are truly septic you might end up giving 10-15 liters. The first response to low BP is always going to be fluids and levophed will be useless without something to squeeze.

2) Never use vasopressin for sepsis. I known that bundles have that order, but in my opinion, vasopressin should be removed from the bundle. The only time vasopressin should be used is when all other vasoactive gtts are maxed out or the pt is in DI. It drives me absolutely insane when I get a septic floor pt who was automatically put on vasopressin. "Well, the pt's urine output basically stopped, so I think he's now in renal failure." No, it stopped b/c you put him on vasopressin. That's the side effect of a continuous gtt of antidiuretic hormone. I immediately wean off the vaso and put up Levo, and like magic, there's output again. I think people like vaso because it's really not much of a titrated gtt. They put it on at .04 and move on. It's easiest.

I get your rationale but we don't actually use vasopressin in sepsis for it's antidiuretic effect, vasopressin also has a direct vasoconstriction effect and is considered to be a "potent" vasopressor. Also, at the standard sepsis dose of 2.4 units/hr, vasopressin has been shown in multiple studies to increase urine output.

Sepsis patients will typically have some level of catecholamine resistance which is why supplementing catecholamine vasopressors with vasopressin can be more effective that just using higher doses of catecholamines. The benefit of adding vasopressin is less significant in patients without any clinically significant catecholamine resistance, but is fairly pronounced in patients with resistance.

"Never use vasopressin for sepsis"

You're kidding right? You're scaring me if this is not a joke. MunoRN is absolutely correct. Someone needs to do a little book learning.