Sepsis Purvin Shah, DO, MS, FCCP, MBA 5/15/2014 15

Purvin Shah, DO, MS, FCCP, MBA
15th Annual CV & Medicine Symposium
1. Introduction
2. Definitions
3. Epidemiology
4. Pathophysiology
5. Evaluation & Management
6. Prognosis
7. Conclusion
Why do we care so much about sepsis?
• Sepsis and its advanced forms (severe sepsis and septic shock):
Are major healthcare problems
Affect millions of people around the world each year
Kill 1 in 4 patients (and often more)
Are increasing in incidence
• Just like MI and Stroke, the outcome of patients with sepsis is highly
dependent on the speed and appropriateness of therapy in the initial
hours after sepsis develops.
• As a non-ICU physician, how do YOU play a role in caring for patients
with sepsis?
• Surviving Sepsis Campaign (SSC) Guidelines:
“…the committee believes that the greatest outcome improvement
can be made through education and process change for those caring
for severe sepsis patients in the non-ICU setting and across the
spectrum of acute care.”
Critical Care Medicine, Feb. 2013, Vol. 41, No. 2, pp 577-637
Surviving Sepsis Campaign (SSC) Guidelines
• 2004 – Initial SSC guidelines were published in 2004 and incorporated
evidence available through the end of 2003
• 2008 – Updated SSC guidelines in 2008 based on evidence available
through the end of 2007
• 2012 – Revised SSC guidelines based on literature through Fall 2012
• Quality of evidence = high (A) to very low (D)
• Strength of recommendations = strong (1) or weak (2)
• Spectrum of severity of illness:
Sepsis/SIRS  Severe Sepsis  Septic ShockMODS
• 1991 – initially defined by ACCP and SCCM
• 2001 – reconsidered by ACCP, SCCM, ATS, ESICM, SIS (Surg. Inf. Soc.)
• 2012 – SCCM and ESICM (European Society of Inten. Care Medicine)
• Infection – presence of bacteria in a normally sterile tissue
• Bacteremia – presence of bacteria in blood
• Sepsis – complicates severe infection
• SIRS – complicates a non-infectious insult
• Severe Sepsis – hypoperfusion or organ dysfunction
• Septic Shock – sepsis-induced hypotension despite fluids
• MODS – Multiple Organ Dysfunction Syndrome
• Sepsis
• Problem with regulating the body’s inflammatory response to an infection
• Characterized by the cardinal signs of inflammation in tissues remote from the infection:
• Vasodilation
• Leukocyte accumulation
• Increased microvascular permeability
• SIRS – Systemic Inflammatory Response Syndrome
• Identical to Sepsis BUT complicates a non-infectious insult:
• Pancreatitis, Surgery, Burns, Autoimmune D/O, Trauma, Thromboembolism, etc.
• Massive and uncontrolled release of pro-inflammatory mediators
leads to widespread tissue injury and possibly MODS (high mortality)
• Diagnostic Criteria for Sepsis = probable or documented infection
PLUS some systemic manifestations of infection:
• General:
Temp. >38.3 C or <36 C
HR > 90/min. or >2 std. dev. above normal value for age
Resp. Rate > 20/min.
Altered mental status
• Inflammatory variables:
• WBC > 20K or < 4K
• Normal WBC with >10% immature forms (e.g. Bands)
• Plasma Procalcitonin more than 2 std. dev. above normal value
• Diagnostic Criteria for Sepsis (continued):
• Hemodynamic variables:
SBP < 90 mmHg
MAP < 70 mmHg
SBP decrease > 40 mmHg, or
SBP < 2 std. dev. Below normal for age
• Organ Dysfunction variables:
Hypoxemia (PaO2 / FiO2 < 300)
Oliguria (urine output < 0.5 mL/kg/hr for >2 hrs. despite fluids)
Platelets < 100K
Plasma total bilirubin > 4 mg/dL
• Diagnostic Criteria for Severe Sepsis = sepsis-induced tissue
hypoperfusion (low BP, high lactate or oliguria) or organ dysfunction
(any of the following thought to be due to the infection):
Sepsis-induced hypotension
Lactate > ULN (upper limits of normal)
U.O. < 0.5 mL/kg/hr. for >2 hrs. despite adequate fluid resuscitation
Acute lung injury (PaO2/FiO2<250) w/o pneumonia as infection source
Acute lung injury (PaO2/FiO2<200) with pneumonia as infection source
Creatinine > 2 mg/dL
Platelet count < 100 K
Coagulopathy: INR > 1.5
• Septic Shock: sepsis-induced hypotension that persists despite
adequate fluid resuscitation (30 mL/kg of crystalloids – some of this
may be albumin)
• MODS: progressive organ dysfunction
• The severe end of the illness spectrum
• Primary MODS – organ dysfunction occurs early and directly from the insult
(e.g. renal failure due to rhabdomyolysis)
• Secondary MODS – organ failure is NOT due to the insult itself but rather is a
consequence of host response (e.g. ARDS in patients with pancreatitis)
• Includes: PaO2/FiO2 ratio, Platelet count, Bilirubin, Creatinine, Low BP, GCS
• Incidence: More than 1,665,000
cases diagnosed in the U.S.
• An analysis reported increased
rates from 13 to 78 cases per
100,000 between 1998 and
• Greatest in African-American
males, during winter and
patients > 65 yrs of age.
• Pathogens: Gram-positive bacteria more than Gram-negative.
Also, fungal sepsis has increased over the past decade.
• Disease Severity: Increase in disease severity is noted – e.g., severe
sepsis increased from 26% to 44% over a 10-year period.
Most common organ dysfunctions: ARDS, AKI, DIC.
• Mortality: High mortality rate (12 to 50%). Increases stepwise:
SIRS = 7%, Sepsis = 16%, Severe Sepsis = 20%, Septic Shock = 46%
Mortality remains elevated at one-year among survivors of sepsis.
Risk Factors for Sepsis
• Bacteremia (study of 270 blood cultures: 95% of positive blood
cultures were associated with sepsis, severe sepsis, or septic shock)
• Advanced age (>65 yrs) – age is an independent predictor of mortality
• Immunosuppression – neoplasm, renal failure, hepatic failure, AIDS,
asplenism or immunosuppressant medications
• Diabetes
• Community-acquired pneumonia (CAP): severe sepsis develops in
48% of patients with CAP
• Normal Response to Infection:
- Host immune cells (esp. macrophages)
recognize and bind to microbial
- Activates an inflammatory response
(release of proinflammatory cytokines
TNF / IL-1, chemokines, NO, etc.)
- Polymorphonuclear leukocytes (PMN)
become activated, migrate to injury
site, release mediators causing
inflammation (edema, erythema,
warmth) and tissue repair & healing.
• Transition to Sepsis:
- the normal inflammatory response
becomes uncontrolled and
generalized because a large quantity
of proinflammatory cytokines are
- Complement cascade gets activated
- Bacterial cell wall components and
bacterial products may enhance
progression to sepsis
• Systemic effects:
widespread cellular injury
derangements in metabolic auto-regulation
imbalances in coagulation and fibrinolytic systems
mitochondrial dysfunction
delayed apoptosis of activated macrophages and neutrophils
• Organ-specific effects:
Hypotension (from diffuse vasodilation), myocardial depression
Pulmonary edema, V/Q mismatch, hypoxemia
Translocation of bacteria and endotoxin in the GI tract
Liver dysfunction
Acute renal failure
Evaluation & Management
• Early Management:
First, stabilize airway & breathing – supplemental O2, monitor
pulse ox, may need to intubate and place on mechanical ventilation.
Check PCXR and ABG.
Then, assess & restore perfusion – if low BP, then insert A-line
catheter for accurate BP. EXAM: HR>90/min, obtundation / restless,
oliguria / anuria, cool vasoconstricted skin. LABS: lactate > 4mmol/L
Insert central venous catheter for infusion of meds & fluids /
blood draws / measuring CVP and central venous O2 sat. (ScvO2).
Evaluation & Management
• Goals of initial resuscitation: (SSC Guidelines) within 1st 6 hours
based on an EGDT trial of 263 patients (and a meta-analysis) with
targeting of specific goals resulted in lower mortality (31% vs. 47%):
- ScvO2 > 70%
- CVP = 8-12 mmHg
- MAP > 65 mmHg
- U.O. > 0.5 mL/kg/hour
Rivers E et al. “Early Goal-Directed Therapy in the treatment of
severe sepsis and septic shock”; NEJM 2001; 345: 1368-1377
Evaluation & Management
• Lactate clearance – targeting a lactate clearance of >10% (versus
targeting ScvO2 > 70%) gives same mortality benefit, hospital LOS,
ventilator-free days, or incidence of multi-organ failure.
• Recommend targeting normalization of lactate level if initially elevated.
• Other targets (besides ScvO2, CVP, MAP, U.O.) – radial pulse pressure,
aortic blood flow peak velocity, brachial artery blood flow velocity,
passive leg raising, etc. can be used to guide fluid resuscitation.
Evaluation & Management
• Screening – recommend routine screening for Sepsis in potentially
infected seriously ill patients (so treatment can be started early)
• Diagnosis – 2 sets of blood cultures BEFORE Abx Tx is started (as long
as there is no significant delay >45 min. in starting Abx).
- blood cultures (peripheral and through each vascular access device)
- sputum cultures
- urine cultures
- wound cultures
- rapid influenza antigen testing (during the flu season)
- appropriate imaging studies (to identify abscess for I&D, etc.)
Evaluation & Management
• Antibiotics – administer effective IV Abx within the 1st hour of
recognizing septic shock and severe sepsis.
• Each hour delay is associated with increase in mortality
• Initial empiric Abx Tx should have activity against all likely pathogens
(bacterial, viral, fungal) and penetrate in adequate concentrations into the
tissues that are thought to be infected.
Hospitalized patients – Gram-positives are most common, then Gram-negatives
Neutropenic patients – wide range of potential pathogens, esp. Gram-negatives
Avoid recently used anti-infective agents
Add fluconazole or an echinocandin if candidemia is a concern
• (e.g., neutropenia, immunosuppressed, prior broad-spectrum abx tx, etc.)
Evaluation & Management
• Antibiotics –
• Studies have shown combination therapy produces better clinical outcomes
in severely ill, septic patients with a high risk of death.
• Duration = 7-10 days based on clinical judgment
• Longer if:
slow clinical response
undrainable focus of infection
bacteremia with S. aureus
some fungal and viral infections
• Re-assess patients daily for de-escalation of abx tx once pathogen is known.
• Start anti-viral tx early if a viral infection is suspected (e.g., influenza).
Evaluation & Management
• Source Control –
• Look for specific anatomical diagnosis of infection and intervene w/in 12 hrs.
• Necrotizing soft tissue infection  debridement
• Abscess  drainage
• peritonitis, cholangitis, intestinal infarction  surgical intervention
• Use the intervention with the least physiologic insult
• e.g. percutaneous rather than surgical drainage of an abscess
• Remove any potentially infected intravascular access devices
Evaluation & Management
• IV Fluids – rapid, large volumes are indicated (e.g., 5 liters in 6 hours)
unless CHF is present. Repeat IV fluid boluses until:
- BP is acceptable
- tissue perfusion is acceptable
- CHF develops
- perfusion fails to improve with IV fluids
• Crystalloids – initial fluid of choice (e.g., NS, LR) = 30 mL/kg bolus
• Albumin (more costly) – should be used if amount of crystalloids will be too much.
• DO NOT USE hydroxyethyl starch (Hespan) b/c of increased mortality (51% vs. 43%).
SAFE trial, 6S trial, VISEP trial
Evaluation & Management
• Vasopressors – used in patients that remain hypotensive despite
adequate fluid resuscitation (or that develop pulmonary edema from
IV fluids).
Target – MAP > 65 mmHg
Norepinephrine is preferred 1st choice (more potent, more effective for BP)
Epinephrine – 2nd choice
Vasopressin – can also be added at low-dose
Dopamine – only for highly selected pts (low risk of arrhythmias, low EF)
• Do NOT use low-dose dopamine for renal protection  does not work.
• Phenylephrine – use if norepinephrine causes arrhythmias, high CO with low
BP or add as combination therapy to other vasopressor/inotrope drugs
Evaluation & Management
• Inotropic Therapy –
• Dobutamine –
• Upto 20 mcg/kg/min added to vasopressors
• Use if:
• Persistent Hypoperfusion
• ScvO2 remains < 70% (despite adequate intravascular volume and adequate MAP)
• Myocardial Dysfunction (low CO) is present
• Recommend against using it to increase cardiac index to a predetermined supranormal
levels – no benefit shown with this strategy
Evaluation & Management
• Corticosteroids –
• Use only if BP is still low despite adequate fluid resuscitation and vasopressor
• Dose of 200 mg / day in divided doses (e.g., 100 mg Hydrocortisone Q12 hrs)
• Do not use the ACTH stimulation test to identity pts to give steroids
• Random cortisol levels have not been demonstrated to be useful.
• Taper steroids when pt is off vasopressors
• No recommendation for optimal duration
• No difference in outcome was noted in 3 vs. 7 days of treatment
• Do not use as “preventive” therapy if shock is not present
Evaluation & Management
• Packed Red Blood Cell (PRBC) Transfusions –
• Give PRBCs when Hb < 7.0 g/dL
• If no MI, ischemic CAD, severe hypoxemia or acute hemorrhage
• Optimal Hb level is not known but Hb b/t 7-9 vs. 10-12 g/dL was NOT
associated with increased mortality in critically ill pts.
• Do not use erythropoietin as a tx of anemia in severe sepsis
• No specific info. is available regarding its use in septic pts.
• Fresh Frozen Plasma (FFP) –
• Do not use to correct clotting abnormalities unless bleeding is present or
invasive procedures are planned
Evaluation & Management
• Platelets –
• Sepsis is thought to be associated with
• Limitation of platelet production
• Increased platelet consumption
• Transfuse when:
• Plts < 10K if no bleeding is present
• Plts < 20K if pt has a significant risk of bleeding
• Higher platelet levels (> 50K) are advised if:
• Active bleeding
• Surgery
• Invasive procedures
Evaluation & Management
• Recombinant Activated Protein C (rhAPC, Xigris) –
• Approved in 2001 after PROWESS trial showed a significant reduction in
mortality (24.7% vs. 30.8% with placebo)
• Shown to be ineffective in later studies (e.g., PROWESS SHOCK trial in 2011)
so the drug was withdrawn from the market by Eli Lilly.
Evaluation & Management
• Glucose Control –
• Recommend a protocolized approach to be used when 2 consecutive blood
glucose levels are > 180 mg/dL.
• Target blood glucose level < 180 mg/dL.
• Monitor blood glucose levels every 1-2 hours.
• Other –
• DVT / PUD Px, Enteral feedings, vent. support
Surviving Sepsis Campaign Bundles
• Within 3 hours:
• Measure lactate level (to assess perfusion)
• Obtain blood cultures before giving antibiotics
• Administer broad-spectrum antibiotics
• Administer 30 mL/kg crystalloid if:
• Hypotension
• Lactate > 4 mmol/L
Surviving Sepsis Campaign Bundles
• Within 6 hours:
• Start vasopressors if hypotension persists despite initial fluid resuscitation
• Maintain MAP > 65 mmHg
• If hypotension persists despite volume resuscitation OR initial lactate > 4
mmol/L, then:
• Measure CVP (central venous pressure) – target CVP > 8 mmHg
• Measure ScvO2 – target ScvO2 > 70%
• Remeasure lactate if initial lactate was elevated
• Target lactate – normal value (< 2.2 mmol/L)
ProCESS Trial
• Protocolized Care for Early Septic Shock – 31 hospitals in the U.S.
• 1341 patients in 3 groups: protocol-based EGDT (439 pts), protocol-based
standard therapy (446 pts), or usual care (456 pts)
• Results: 60-day mortality rates were similar (21.0%, 18.2%, 18.9%). No
significant differences in 90-day or 1-year mortality.
• SSC Response: usual care mortality (18.9%) is already low (b/c of overall
improvement in our care of these sick pts – 70% of hospitals use a “sepsis
protocol”) vs. 46.5% in 2001 EGDT trial. Also, this study does not address
pts with severe sepsis without septic shock – aggressive mgt of these pts
has likely lowered mortality rates.
“A Randomized Trial of Protocol-Based Care for Early Septic Shock”.
NEJM May 1, 2014, Vol. 370, No.18, pp. 1683-1693
Other Recent Trials
• High vs. Low Blood-Pressure Target in Patients with Septic Shock
(NEJM 4/24/14, vol. 370, no. 17, pp. 1583-1593) – 776 pts
• No significant difference in mortality (28-day or 90-day) with MAP of 65-70
mmHg vs. 80-85 mmHg
• Albumin Replacement in Patients with Severe Sepsis or Septic Shock
(NEJM 4/10/14, vol. 370, no. 15, pp. 1412-1421) – 1818 pts
• Use of Albumin in addition to crystalloids did not improve survival at 28 and
90 days.
Prognostic Factors
• Host Response – Leukopenia or Failure to develop a fever = poor prognosis
• Co-Morbidities – new-onset a. fib., age > 40 yrs, AIDS, liver disease, cancer,
EtOH abuse, immunosuppression = poor prognosis
• Site of infection – UTI = lowest mortality rate BUT GI or pulmonary source
= higher mortality
• Type of infection – nosocomial infection = higher mortality (compared to
community-acquired pathogens)
• Abx Tx – Appropriate Abx Tx = lower mortality. BUT, prior Abx Tx = higher
• Failure to restore perfusion early = higher mortality
• Sepsis –
Is a major healthcare problem with
an increasing incidence and
a high mortality rate
affecting millions of people around the world every year.
• Its outcome depends on the speed and appropriateness of therapy early after
recognizing its presence.
• There are evidence-based guidelines published to guide its management with
updates every 4 years.