Transcript Sepsis - Dr Qureshi

Severe sepsis and septic shock
Salman Qureshi, MD
McGill University Health Centre
July 7, 2010
[email protected]
Severe sepsis and septic shock
• Estimated 750,000 cases per year
–
–
–
–
Projected to increase 25% by 2020 (1.1 million cases per year)
Average age: 6th decade
50% are admitted to ICU
15% require mechanical ventilation
• Overall mortality of 30-60%
– 10th leading cause of death in USA
– Most common cause of death in non-coronary ICU
• Annual expenditure of 17 billion dollars
Crit Care Med 2001 29:1303
Clinical Vignette I
•
•
•
•
•
•
40 year old African female living in Montreal
Flu-like symptoms for 2-3 days
Fever, generalized malaise
Onset of sharp right posterior thoracic pain
Presents to outpatient clinic
T=40°C, P=110/min, BP 80/50, R=35/min
Clinical Vignette II
•
•
•
•
•
88 year old female; excellent health prior to admission
Seen in ER for leg pain, swelling
Diagnosed with DVT and cellulitis
Admitted to SSU; treated with IV Cefazolin and LMWH
Respiratory distress, severe hypoxemia, fever, rapid
wide complex irregular heart rhythm with profound
hypotension
Management questions
• What diagnostic tests would be most helpful?
• What is the correct choice and timing of antibiotics?
• How much colloid or crystalloid should we use to
resuscitate the patient?
• How will the adequacy of resuscitation be evaluated?
• Should we use adjunctive therapies?
–
–
–
–
Catecholamines or vasopressin
Low-dose corticosteroids
Intensive insulin therapy
Activated protein C
• What complications should be anticipated?
Sepsis- what is it?
Sepsis is a syndrome
• Current definitions encompass an extraordinarily
heterogenous mix of patients:
– Different underlying co-morbidities
– Variable host immune/inflammatory capabilities
– Different types, quantities, and virulence of
microorganism (gram positive vs. gram negative
bacteria vs. fungal, etc.)
– The type and pattern of organ involvement (lung vs.
blood vs. peritoneum, etc.)
Incidence of sepsis
N Engl J Med 2003 348: 1546
Microbial etiology of sepsis
N Engl J Med 2003 348: 1546
Microbial etiology of sepsis
• *Gram-positive bacteria 52.1%
– Staphylococcus aureus, Streptococcus pneumoniae
• Gram-negative bacteria 37.6%
– Escherichia coli, Klebsiella pneumoniae, Pseudomonas
aeruginosa
• Polymicrobial infections 4.7%
• Anaerobes 1.0%
• *Fungi 4.6%
• *Greatest relative increase
Need for faster and more precise
diagnostic tests
• 20-30% of patients with severe sepsis/septic shock have
negative blood cultures
• Yet, an infection is causal in most of these cases:
– Adults have intermittent, low magnitude bacteremia
(1-30 CFU/ml) requiring culture of relatively large
blood volumes
– Blood cultures often sterilized by single dose of
antibiotic
– Some microorganisms are fastidious
– False positive results may result from contaminating
skin flora
Real-time PCR for bacteremia
• Rapid detection of microbial DNA
– Sensitivity: 96.2%
– Specificity: 100%
– Limit of detection: 40 cfu/ml
– Turnaround time: 4 hours
• Uses universal primers designed to amplify conserved
stretches of DNA from any bacterium
• Confirmation of microbial species by nucleotide
sequencing is required
Exp Biol Med 2005 230:587
J Mol Diag 2005 7:581
Searching for an ideal sepsis marker
• Characteristics:
–
–
–
–
–
Rapid diagnosis
Stage and prognosticate the disease
Differentiate infectious from non-infectious causes
Differentiate viral from bacterial from fungal infection
Reflect effectiveness of antimicrobials and source
control
– High sensitivity and specificity
– Easy to perform
– Low cost
The host response
When we sense lipopolysaccharide, we are likely to turn on
every defense at our disposal; we will bomb, defoliate,
blockade, seal off and destroy all tissue in the area…
Pyrogen is released from leukocytes, adding fever to
hemorrhage, necrosis and shock. It is a shambles…
All this seems unnecessary, panic-driven.
There is nothing intrinsically poisonous about endotoxin, but it
must look awful, or feel awful, when sensed by cells.
-Lewis Thomas
N Engl J Med 1972 287: 553
The host response
• Severe sepsis and septic shock are thought to be
caused by a systemic “dysregulated” inflammatory
process.
• At present we cannot rapidly measure the patient’s
ability to produce an appropriate inflammatory response,
as opposed to an excessive or inadequate response.
• We also cannot readily measure the adequacy of
response nor do we have the tools to interpret this
information rapidly and then to appropriately modulate
this response to the benefit of our patients.
Crit Care Med 2008 36: 964-966
Immunopathogenesis of prolonged
sepsis
Lancet Infect Dis 2001 1:165
Current paradigm for immunologic
response during sepsis
N Engl J Med 2003 348:138
Infect Dis Clin N Am 2009 23: 485-401
Surviving Sepsis Campaign
• Early goal directed resuscitation during the first 6 hours after
recognition (1C)
• Administration of broad-spectrum antibiotic therapy within 1 hr of
diagnosis of septic shock (1B) and severe sepsis without septic
shock (1D)
• Recombinant activated protein C with severe sepsis and clinical
assessment of high risk for death (2B except 2C for postoperative
patients)
• Stress-dose steroid therapy given only in septic shock after blood
pressure is identified to be poorly responsive to fluid and
vasopressor therapy (2C)
• Vasopressor preference for norepinephrine or dopamine to maintain
an initial target of mean arterial pressure >65 mm Hg (1C)
Crit Care Med 2008 36:296
Early goal-directed therapy
• Global tissue hypoxia increases during severe sepsis
and septic shock and precedes multi-organ failure and
death
• Goal-directed therapy provided at the earliest stages of
severe sepsis and septic shock restores the balance
between oxygen supply and demand
• Reduced in-hospital mortality (30.5% vs 46.5%)
P=0.009
– More fluid, RBC, inotropes in first 6 hours for EGDT
– More fluid, RBC, vasopressors, mechanical ventilation, PA
catheters (7-72 hours) for standard therapy group
N Engl J Med 2001 345:1370
N Engl J Med 2001 345:1370
Chest 2007 132:425
“In this investigation we demonstrate…a 9% absolute and 33%
relative mortality reduction of implementing EGDT in ED patients
with severe sepsis and septic shock. Our data suggest a number
needed to treat of approximately 11 persons.”
Delay in antimicrobial therapy and
septic shock mortality
• Retrospective review of 2731 septic shock cases
• Persistent or recurrent hypotension (MAP <65 mmHg,
SBP <90 mm Hg, or drop in SBP >40 mm Hg) despite
>2L fluid
– Average age 62.7 years; 54.3% males
– Overall mortality 56.2%
– 77.9% had documented infection
• Over the first 6 hours after onset of hypotension, each
hour delay in administration of effective antibiotic therapy
was associated with a mean increase in mortality of
7.6%
• Multivariate analysis showed that time to antimicrobial
initiation was the strongest predictor of outcome
Crit Care Med 2006 34:1589-1596
Septic shock mortality and
antibiotic initiation
Average time to antibiotic implementation: 13.51±0.45 hours
Median time to antibiotic implementation: 6 hours
Crit Care Med 2006 34:1589
Subgroup analysis
Crit Care Med 2006 34:1589
Appropriate antimicrobial therapy
and septic shock mortality
Infect Dis Clin N Am 2009 23: 485-401
Initial antibiotic selection for severe
sepsis and septic shock
Infect Dis Clin N Am 2009 23: 485-401
Septic shock: The golden hour
• In the clinical context, a “golden hour” has been
described for therapy of hypovolemic shock due to
trauma, cardiogenic shock due to myocardial infarction,
and, most recently, obstructive shock due to massive
pulmonary embolus
• In serious human infections, rapidity of antimicrobial
therapy following presentation has been understood to
be a critical determinant of outcome for specific
conditions including community-acquired pneumonia,
ventilator-associated pneumonia, meningitis, bacteremia,
and now, septic shock
Vasoactive therapy
• Tissue perfusion requires both pressure and flow
– Lactic acidosis is a relatively late marker of inadequate tissue
perfusion
• Vasopressors: increase vascular resistance
– Pressure target: MAP >65 mm Hg or SBP >90 mm Hg
• Inotropes: increase cardiac contractility
– Flow target: Cardiac index >2.2 L/min/m2 or CI that results in a
central venous saturation >70% or mixed venous saturation
>65%
• Vasoactive therapy follows adequate volume expansion
Catecholamines I
• Norepinephrine
– Strong a1 agonist; weak b agonist; initial agent of choice for low
BP in septic shock
– 1-20 ug/minute IV
• Phenylephrine
– Pure a agonist; does not induce tachycardia
– 20-200 ug/minute IV infusion; 100-200 ug IV bolus lasts ~2-3
minutes
• Epinephrine
– Potent b1> b2 agonist, a1 agonist; arrythmogenic; associated
wtih lactic acidosis
– 1-10 ug/minute IV
Catecholamines II
• Dopamine
– Chemical precursor of epinephrine and norepinephrine
– Dose dependent renal/mesenteric vasodilation, b1,
a1 effects; arrythmogenic
– 2.5-10 ug/kg/minute IV infusion
• Dobutamine
– Inotrope; raises heart rate and contractility
– b1 and b2 agonist with variable effect on BP
– 2.5-10 ug/kg/minute IV infusion
Milrinone
• Non-catecholamine
• Potent stimulator of cardiac contractility
– Improved right and left heart function
• Pulmonary > systemic vasodilation
– Lowers pulmonary artery pressure
– Systemic hypotension
• Loading dose 50 ug/kg bolus IV
• Infusion 0.25-0.75 ug/kg/minute IV
• Duration of action: 4 hours
Vasopressin
• Non-catecholamine; stimulates specific receptors
• Useful as catecholamine-sparing agent in refractory
hypotension
• Mesenteric vasoconstriction at higher dose
• 0.01-0.04 Units/minute IV infusion
• Up to 40 Unit bolus IV in ACLS protocol
Vasopressin and septic shock trial
(VASST)
• Randomized, multicenter, controlled trial involving 778
patients with septic shock
– Evaluation of low-dose vasopressin (0.01 to 0.03 U per minute)
added to norepinephrine vs. norepinephrine alone
• No difference in the primary end point of 28-day mortality
(35% and 39%, respectively; P = 0.26) or 90-day
mortality (43.9% and 49.6%, respectively; P = 0.11)
• Patients with less severe septic shock (those with a
requirement for 5 to 14 μg of norepinephrine per minute
at baseline) had a significant reduction in mortality with
vasopressin therapy (26.5% vs. 35.7%, P = 0.05)
N Engl J Med 2008 358:877
VASST: caveats
• Low overall mortality; high risk patients excluded (acute
ischemic heart disease or CHF)
• Catecholamine alone achieved a mean BP of 72 mm Hg
– Study participants were not truly refractory to
catecholamines
– Vasopressin was used a catecholamine sparing agent
• Average time from meeting entry criteria and infusion of
study drug was 12 hours
– past the window for early goal-directed therapies
Sepsis: a prothrombotic state
Host defense that aims to contain microbes/inflammation
Uncontrolled systemic microvascular thrombosis contributes to multiple organ dysfunction
N Engl J Med 1999 340: 207
Activated Protein C (APC):
anticoagulant properties
N Engl J Med 2001 344:699
Activated Protein C
• An endogenous protein that is converted from it inactive
precursor by an endothelial thrombin-thrombomodulin
complex
• APC levels fall during severe sepsis; associated with
increased mortality
• Inhibits thrombosis: Antagonizes factor Va and VIIIa
• Promotes fibrinolysis: Inhibits expression of plasminogen
activator inhibitor-1
• Anti-inflammatory: Reduces monocyte cytokine release
Randomized trials of APC in sepsis
• PROWESS trial (N Engl J Med 2001 344:699)
– 6.1% absolute mortality reduction for sepsis-induced
organ dysfunction associated with a clinical
assessment of high risk of death (APACHE II ≥25 or
multiple organ failure)
• ADDRESS trial (N Engl J Med 2005; 353:1332)
– Not recommended for adults with a low risk of death
(APACHE II<20)
• ENHANCE open-label trial (CCM 2005 33:2266 )
– Some evidence that earlier administration (<24h) is
beneficial
Adrenal Insufficiency
• Waterhouse R. Case of suprarenal apoplexy. Lancet
1911;1:577
• Large randomized trials of high dose corticosteroid
therapy have been negative
• In critical illness, a state of relative adrenal insufficiency
may develop that is associated with persistent
catecholamine dependence
• Cortisol is a permissive factor for catecholamine action
Am J Respir Crit Care Med 2006 174:1319
Prognostic Stratification
Baseline cortisol
250 ug cosyntropin
28 day mortality
(nmol/L)
(ACTH)
< 940
> 250
26%
> 940
< 940
> 250
< 250
67%
> 940
< 250
82%
Overall mortality 58% (51-65)
Median survival 17 days (14-27)
JAMA 2000 283: 1038
Low dose steroid trial
• 300 patients in 19 ICUs in France
• Cosyntropin (ACTH) test performed on all
• Randomized to placebo vs:
• 50 mg hydrocortisone iv q6h x 7days
• 50 ug fludrocortisone po od x 7 days
• 7% absolute reduction in mortality among ACTH nonresponders treated with steroids (p=0.02)
• No significant difference in 28 day mortality in ACTH
responders: trend towards increased mortality
JAMA 2002 288: 262
Corticosteroid Therapy of Septic
Shock (CORTICUS) study
• 499 patients with septic shock and hypotension for >1
hour after adequate fluid resuscitation
• Randomized to hydrocortisone 200 mg/d for 5 days with
tapering over 6 days vs. placebo; no fludrocortisone
• Findings:
– Faster reversal of shock in hydrocortisone group
– No overall mortality difference at 28 days (RR of death 1.09;
95% CI 0.84-1.41) regardless of cosyntropin response
– Increased incidence of superinfection, including new episodes of
sepsis or septic shock in the hydrocortisone group
• Only 35% power to detect 20% mortality difference
N Engl J Med 2008 358:111
Controversy over cortisol
• Free cortisol, rather than the protein-bound fraction, is
responsible for the physiologic function of the hormone
• More than 90 percent of circulating cortisol in human
serum is bound to proteins
– corticosteroid- binding globulin: high affinity low capacity
– albumin: low affinity high capacity
• Alterations in the binding proteins could affect measured
concentrations of serum total cortisol and, thus, the
interpretation of tests used to assess adrenal function
N Engl J Med 2004 350:1629-38
Intensive insulin therapy
• Tight glycemic control (4.4-6.1 mmol/L) conferred a 3.4%
absolute mortality benefit among patients in a surgical
intensive care unit (N Engl J Med 2001 345:1359)
– 42% relative mortality reduction among patients in ICU >5 days
– Single centre study with liberal use of parenteral calories
– Significant increase in hypoglycemia <2.2 mmol/L
• Follow-up study in 3 medical ICUs did not show an
overall mortality benefit (N Engl J Med 2006 354: 449)
– Subgroup analyses demonstrated less acute renal injury and
shorter duration of mechanical ventilation
– Mortality benefit for patients in ICU >3 days, but difficult to
accurately predict who will be in this group at ICU admission
VISEP: Efficacy of Volume Substitution
and Insulin Therapy in Severe Sepsis
• Terminated early; no mortality benefit or difference in
organ failure scores at 28 or 90 days
– Severe hypoglycemia was reported in 17.1% of the
intensive-therapy group and 4.1% of the
conventional-therapy group (P<0.001)
– Hypoglycemia was an independent risk factor for
death from any cause- but could be a marker of poor
outcome rather than causal
N Engl J Med 2008 358:125
NICE-SUGAR trial
• Normoglycemia in Intensive Care Evaluation and
Survival Using Glucose Algorithm Regulation)
• Large open-label RCT (6104 patients) comparing insulin
therapy targeting glucose range of 4.5-6.0 mmol/L vs.
8.0-10.0 mmol/L
• Higher mortality at day 90 in intensive glucose control
group (27.5% vs 24.9%; p=0.02)
• No significant differences in ICU or hospital LOS or rates
of organ dysfunction
• Current recommendation is a target glucose range of
8.0-10.0 mmol/L in ICU
N Engl J Med 2009 360:1283-97
Clinical Vignette I
•
•
•
•
•
•
40 year old African female living in Montreal
Flu-like symptoms for 2-3 days
Fever, generalized malaise
Onset of sharp right posterior thoracic pain
Presents to outpatient clinic
T=40°C, P=110/min, BP 80/50, R=35/min
Clinical vignette I: outcome
•
•
•
•
•
•
•
Microbial etiology: positive sputum but negative blood Cx
Antibiotics: broad-spectrum I.V. within one hour
Resuscitation strategy: EGDT with crystalloid
Glycemic control: Insulin to target glucose 8-10 mmol/L
Vasopressors: norepinephrine alone
Activated Protein C: No (platelet count <30,000)
Corticosteroids: No (good response to pressors)
Clinical Vignette II
•
•
•
•
•
88 year old female; excellent health prior to admission
Seen in ER for leg pain, swelling
Diagnosed with DVT and cellulitis
Admitted to SSU; treated with IV Cefazolin and LMWH
Respiratory distress, severe hypoxemia, fever, rapid
wide complex irregular heart rhythm with profound
hypotension
Clinical vignette II: outcome
•
•
•
•
•
CXR, EKG, troponin, blood cultures
IV ticarcillin-clavulanic acid plus vancomycin
Intubation/mechanical ventilation
Early goal directed therapy
Swan-Ganz catheter; titration of norepinephrine,
epinephrine, milrinone to CI>2.2 and MvO2>65%
• Low-dose corticosteroids
• Glucose control (8.0-10.0 mmol/L)
• Consideration of recombinant activated Protein C
Summary:
Severe sepsis and septic shock
Effective management requires a coordinated approach to diagnosis
and treatment
Multiple evidence-based interventions are available
EGDT
Vasoactive therapy
Corticosteroids
APC
Insulin
Low tidal volume ventilation
Antibiotics/Source Control/Ulcer and DVT prophylaxis
Time is of the essence