I just wanted to post some of the information we covered about sepsis in case you wanted to give it a second look. Some good references are linked throughout the text. Here is a link for current guidelines- click here
Why do we care?
Mortality rate is
around 1/3 with severe sepsis
SIRS (2/4):
1.
HR > 90
2.
Temp. < 36
, > 38
4.
WBC > 12
000 or < 4000
Sepsis:
Inflammatory
reaction (SIRS) + probable evidence of infection
Severe Sepsis:
Sepsis + plus
signs of hypoperfusion. i.e hypotension, oliguria ( < .5 ml/kg/hr), elevated
lactate, elevated creatinine, elevated bilirubin, hypoxia
Septic Shock:
Sepsis + signs of
hypoperfusion despite adequate resuscitation
So how do we
resuscitate?
Based on Rivers
2001, NEJM (Rivers paper link)
·
RCT in one
centre in Detroit, sample 263 with severe sepsis or worse
·
Randomized to
early goal directed therapy v. usual care
·
Mortality
46.5 v. 30.5 %
·
Attempted to
meet certain benchmarks within 6 hours of admission
1. Inserted lines
2. CVP 8 -12 mm Hg with fluids (CVP is
pressure in SVC near R. atria = JVP + ~5 cm)
3. MAP > 65 mm Hg with vasopressors
preserves perfusion (MAP = 2/3 diastole + 1/3 systole)
4. Scv02 < 70 %? (N.B Superior
vena cava 02 saturation), check hematocrit!
Scv02 is a measure of how much oxygen is in
blood after passing through body, so it could be low to either not enough
delivery or too much consumption. We need to ensure enough delivery of oxygen
to tissues.
a. < 30 %, transfuse pRBCs
b. > 30 %, use inotropic agents (in this
case dobutamine)
·
So bottom
line: using a protocol with hard outcomes works!
We try to do this,
also target:
1.
Urine output
> .5 ml /kg/hr
2.
Normalizing
lactate, can be used instead of ScvO2 , target >20% decrease in 2
hrs
Now we are
thinking about resuscitation, what else should we be doing?
Treating the
underlying infection!
1.
Diagnosis à head to
toe: encephalitis, meningitis, ENT, influenza, pneumonia, GI, GU,
hepatobiliary, skin, deep tissue, osteomyelitis, heart
·
Hx / Physical
·
Imaging
·
Cultures
(minimum 2 sets, peripheral and central)
·
Treat (within
one hour is optimal): broad spectrum, generally combination therapy because
better outcomes in the sickest patients (due to resistance?)
·
N.B: each
hour delay increases mortality by 8%
Some Issues:
1.
What fluid
should we use?
·
Remember
Interstitial ¾ + Vascular ¼
·
Normal saline
is salt water, it moves from intravascular to extravascular space according to
Starling Forces: hydrostatic and oncotic pressure
·
Colloids are
molecules that are too big to cross semi permeable membrane from vessels to
interstitial space
·
Crystalloids:
normal saline (Na 154 , pH 5.5), Ringer’s lactate (Na 131, lactate, K, Ca, pH
6.5)
·
NS is acidic
because the ions decrease the concentration of CO2 in blood.
·
Colloids:
Albumin, Hydroxyethyl starch (HES)
·
Albumin has
been shown to be SAFE (see SAFE study, link) in comparison to crystalloids, and may
be better in severe sepsis (not a primary outcome)
·
HES have been
shown to not offer much benefit compared to crystalloids and may cause harm
(see CHEST study, link) as it leads to more RRT (renal replacement therapy)
2.
Which
pressors should we use?
·
Norepinephrine
is first choice (mainly vasoconstriction) compared to dopamine in
meta-analysis, 2-40 mcg/min
·
Epinephrine
is a great adjunct as no evidence of worse outcomes (theoretical splanchnic
vasoconstriction), 1-20 mcg/min
·
Dobutamine is
the first line ionotrope for low cardiac output
3.
Should we use
corticosteroids?
·
Not in the
initial resuscitation phase
·
Evidence is
mixed. Major trial is CORTICUS (link): no mortality benefit, but not only patients
non-responsive to pressors, and probably a less sick cohort
·
If
resuscitation resistant shock trial IV hydrocort 200 mg/day
·
ACTH
stimulation test not useful for deciding who to treat
4.
What is our
hemoglobin level?
·
Our goal is a
hb of 70, based on TRICC trial (link): transfusion level of 70 v 100 with no
difference in mortality.
5.
Should we
have tight glucose control?
·
We should aim
for blood sugars between (10 - 6) to avoid hypoglycemia
·
Initial trial
(leuven protocol) found reduced mortality with tight control (4-6)
·
NICE-SUGAR
trial (link)found increased mortality with tight control from hypoglycemia
·
Subsequent
meta-analyses have differed, may be due to sugar level of controls in each
study
6.
Should we be
using bicarbonate?
·
No if pH is
> 7.15, as this has been studied in two small RCTs
·
Unclear if
should be used in pH < 7.15. Remember bicarb will complex with hydrogen to
make CO2 and H2O, and CO2 can enter cells and
cause worsening acidosis