Sepsis

    Sources:  See "References"

    Introduction & Definitions

    • Severe catabolic reactions:
      • Widespread endothelial reaction
      • Release of inflammatory agents
      • Resulting in a deficit of oxygen delivery
    • Complicated by: Multi-organ failure (70-90%)
    • Mortality: 15-60%, higher with comorbidities.
      • Add 15-20% mortality for each sepsis-induced organ dysfunction.
    • Mechanism:
      • Triggered by infection
      • Activate many cytokines (i.e. TNF-alpha, IL-1), activate leukocytes (adhesion, degranulation) --> endothelial damage.
      • Causes vasodilatation, capillary leak, decreased effective circulating volume.
        • Eventually myocardial dysfunction, exacerbates hypotension.
    • Definitions:
      • Systemic Inflammatory Response Syndrome (SIRS)
        • Systemic Inflammatory Response Syndrome..

          Two of:

          1. Temp <36 or >38
          2. Heart Rate >90
          3. WBC <4 or >12  (or >10% bands)
          4. Resp Rate >20/min (or PaCO2 < 32

           

        • Can be due to inflammation or infection.
      • Sepsis:
        • Systemic Inflammatory Response Syndrome (SIRS) caused by microbial products.
        • Sepsis defined by SIRS with suspected infection.
      • Severe Sepsis
        • SIRS caused by microbial products + organ dysfunction.
          • I.e. (Systemic Effects) hypotension, reduced urine output, or metabolic acidosis
      • Septic Shock
        • Organ hypoperfusion despite adequate fluid resuscitation (i.e. unresponsive to volume replacement)
        • Use of vasopressor agents to maintain BP.
    • Bacteremia does not always cause sepsis, and not all sepsis is caused by bacteremia.

    Mechanism

    • Endotoxins
      • Gram-negative Bacteria:
        • Endotoxin is a component of the outer cell wall, and is implicated in the sepsis syndrome.
        • Evidence that Endotoxin is primary substance for sepsis.
          • In particular - the Lipid A portion of the Endotoxin.
          • Lipid A is the outer membrane anchor for the endotoxin molecule, and is thought to be the component causes sepsis.
          • Infusion of Lipid A into animal causes sepsis-like picture.
      • Gram-positive Bacteria:
        • Lipoteichoic acid - mimics endotoxin in Gram negatives.
        • Peptidoglycans
      • Other:
        • Bacterial flagellins, fungal cell walls, and few other components can also lead to sepsis-like picture.
    • Exotoxins
      • Staph aureus - Toxic Shock Syndrome Toxin 1 (TSST-1)
        • Fever, profound shock, conjunctival and palmar hyperemia, desquamation.
      • Strep pyogenes - Streptococcal Pyrogenic Exotoxin A (SPEA)
        • If necrotizing fasciitis can cause shock.
    • Superantigen - bypasses macrophages and activates T-cells directly.  
    • Monocyte-macrophages or dendritic cells respond to endotoxin (LPS)
      • Endotoxin binds TLR (Toll Like Receptors)
        • Proinflammatory cytokines, TNF alpha, interleukin-1, ---> inflammation.
        • Toxic oxygen byproducts.
        • Products that activate complement and coagulation cascades.

    Sepsis Clinical Picture

    • Fever:
      • First and most common manifestation.
      • The higher the fever, the higher likelihood of bacteremia.
      • Hypothermia--> poor prognosis (cannot mount a good immune response)
    • Hemodynamic changes
      • Tachycardia - Concominent with fever.
      • Hypotension
        • Most important outcome determination
        • Failure to reverse early pre-shock leads to irreversible organ damage and death.
    • Acid-base Disturbances
      • Hyperventilation - respiratory alkalosis (due to anaerobic metabolism --> lactic acid buildup)
        • First sign of impending shock.
        • DIAGNOSTIC! even in stage where hemodynamic changes reversible with fluids.
    • Respiratory Changes
      • Tachypnea -
        • Cytokine stimulation of CNS
        • Increased body temperature.
        • Accumulation of lactic acid.
      • ARDS can develop (via endotoxin)
        • endotoxin activates neutrophils that are trapped in small vessels of lungs -- vessel wall damage, leakage of fluid into alveoli.
        • Can mimic pneumonia.

    End Organ Damage Markers

    • Hypoperfusion (lactate, BP)
    • Can Look for End-organ damage:
      • Brain – Decr LOC
      • Kidney – Incr Cr, poor urine output (less than 0.5cc/kg/hr)
      • Liver – Transaminases
      • Lungs – ARDS
      • Cardiac – Trops
      • Intestine – Lactate (anion gap acidosis)
      • Blood - DIC

     

    Investigations

    • In addition to basic bloodwork, CXR, ensure you order:
    • Cultures:
      • Blood cultures X2 sites
      • Urine culture
      • Sputum culture (if chest Xray abnormal)
    • Coagulation profile (INR, fibrinogen, D-dimers or fibrin split products)
    • Blood gasses (+metabolic panels)

    Diagnosis

    • See "Classification"
    • "If fever, tachycardia, and tachypnea with or without leukocytosis are used to define SIRS, then this definition includes other causes in addition to infection.  Therefore active infection must be sought."
    • NOTE: Many cases of pneumonia fit under sepsis picture, but don't require intensive care.
    • Need to find infectious source.
      • Should include positive blood culture (preferrably 2 separate sites), and exclude most cases of coag-negative staph (skin contaminants).
    • Look for organ damage --> defines severity of sepsis.
    • Sources:
    • Source Prevalence
      Lung 35%
      Abdo 21%
      Urinary 13%
      Skin/Soft Tissue 7%
      Others ....

     

    Treatment

    • GOAL:
      • Treat Infection
      • Optimize Tissue Perfusion & OXYGEN DELIVERY

    Risk Stratify

    • High mortality rates:
      • Pseudomonas aeruginosa
      • Candida
      • Polymicrobial bactermia
    • Empiric therapy: need to know local susceptibility patterns!
    • If petechial skin rash
      • meningococcemia
      • gonococcemia
      • S. aureus bacteremia (Localized S.aureus infection)
      • streptococcal bacteremia (Localized S. pygenes infection)
    • Use therapy for most probable site of origin:
    • NOTE: respiratory S.aureus and Enterococcus sepsis unlikely to be life-threatening.

     

    General Principles of Treatment

    • Key is to maintain organ perfusion
    • Sepsis Management Principles

       

      1. 1. PROMPT ANTIBIOTICS! (See Table below for empiric therapy)
      • Start empiric antibiotics within 1 hour and consider changing in 72hrs after culture results.
      • For every hour of delay, mortality goes up 6%
      1. 2. IV FLUIDS!
        • Duration of hypotension before abx is prognostic factor.
        • Be early and aggressive (Each hour delay up to 6h is 7.9% mortality)
        • The idea is to achieve hemodynamic stability and improve tissue perfusion... How to do that:
          • Subject of debate.  Many protocols have been trialed on how to manage severe sepsis in first 6 hours (golden period).  (see below - Rivers, ProCESS Trials).
          • Bottom line:
            • GIVE FLUIDS: Often 6-7L behind, start 500cc q30min to affect BP.
              • Generally need 4-6L in first 6 hours
            • Give NS or Ringers Lactate (colloids equivalent, more expensive)
            • (MOST COMMON ERROR is underestimating volume deficit)
            • May need vasopressors if hypotension persists --> (start norepinephrine)
            • Follow recent evidence and hospital/ICU guidelines.
      2. 3. SOURCE CONTROL!
        • May need surgery consult if collection found.
        • Debride soft tissue infections (Gas in soft tissues?)
        • Any intra-vascular catheter must be removed and cultured.
      3. 4. STEROIDS? -> Controversial (see below)
      4. 5. Drotrecogin alpha? -> Controversial (See below)

     

    • Empiric Antibiotics:
      • Source of infection should be identified and controlled (i.e. removal of infected devices, drainage of abscesses)
      • Blood and source cultures should be obtained BEFORE antibiotics
      • Start empiric antibiotics based on expected site of action:
        • Use of appropriate abx: 17% mortality
        • Use of inappropriate abx: 42% mortality
      • Principles of selecting empric antibiotics:
        • Ensure covers suspected organisms
        • Ensure penetrates site
        • Ensure patient can tolerate it
        • For Pseudomonar consider combination (in US, b/c high level of resistance)
        • For febrile neutropenias consider combination
      • Examples:
        • Levoflox 750mg give now OD
        • Others:
          • Pip-tazo 4.5g IV
          • +vanco 1g IV if MRSA suspected
          • Azithromycin 500mg IV of pneumonia suspected
      • Duration: 7-10 days typically, but longer if slow response or cannot source control or immune deficiency.

     

    Early Goal Directed Therapy

    • Further reading: 2008 Surviving Sepsis Guidelines
    • EARLY GOAL-DIRECTED THERAPY
      • Based on Rivers trial.  Randomized to Rivers protocol vs. standard care.
      • Rivers Protocol:
        • CVP 8-12
          • Give fluid boluses to achieve this. (crystalloid or colloid - equivalent).
          • Most use NS or Ringers Lactate
          • Colloids = Crystalloids (colloids more expensive), but may have an advantage in low oncotic pressure with lots of 3rd spacing. (Liver disease, burns)
        • MAP > 65
          • If non-responsive to fluid challenge, use of pressors if needed (i.e. norepinephrine/dopamine, 2nd line vasopressin)
        • Hb > 90 (transfusion if needed) [Hct >30%]
        • ScvO2 > 70% (Central venous oxygenation)
          • Marker of tissue perfusion: later studies: can also use lactate clearance
          • Suggested in 2008 Surviving Sepsis Guidelines
          • If not achieved, target hematocrit >30% (pRBC)
            • If still low after transfusion, start inotropic agents to improve cardiac contractility (i.e. Dobutamine)
            •  
        • Antibiotics given within 1 hour of presentation
      • Rivers protocol as summarized in the PROCESS Trial:
        • cvp.jpg
      • Monitor:
        • Lactate until back to normal.
        • U/O of ≥0.5 cc/kg/h
        • MAP ≥ 65mmHg
        • SvO2 ≥ 70%
      • Criticism of Rivers trial:
        • 1. Unclear which component gives mortality benefit.
        • 2. Rivers was investing in CVP transducer company (conflict of interest).

     

    • MONITORING
      • Hourly vitals and urine output.
      • Two-hourly ABG's
      • Blood lactate and coags initially and q4-6h (then reassess)
    • Worsening signs:
      • Persistant fall in BP
      • Accumulation of lactate
      • Increasing hypoxemia
      • Coagulopathy

     

    New Evidence - Jones et al. Trial 2010

    • "Lactate clearance vs. central venous oxygen as goals of early sepsis therapy: a randomized clinical trial" Jones et al, 2010.
    • Shows that EGDT based on serum lactate measurement is non-inferior to the EGDT that used ScvO2 monitoring.
    • Both appear to be equal markers of tissue perfusion.

     

    New Evidence - ProCESS Trial

    • PROCESS TRIAL - 2014
      • Randomized blinded multi-center control trial 1341 patients.  (Rivers was single-center and 263 pts)
      • Compared:
        • 1. Protocol-based Early Goal Directed Therapy (EGDT) from Rivers trial (See figure Above)
        • 2. "Protocolized Standard Care"
          • This is less aggressive than EGDT.  ProCESS investigators designed the protocol based on review of literature, two independent ER and ICU physician surveys, standards of practice worldwide and consensus feedback from investigators.
          • NOTE: This protocol did not necessitate a central venous line.  It was placed only if peripheral access was poor.  
          • Target was Given fluids/vasoactive agents to target sBP < 100 and shock index (ratio of HR and sBP) ≥ 0.8.   Fluids were given until team leader decided they were replete.
          • Contrast to EGDT, pRBC were only given if Hb was <7.5 g/dL (<75 g/L)
        • 3. Usual Care
          • Study coordinator collecting data, but not prompting any actions.
          • Lead investigators could not serve as treating physician in usual care group.
        • EGDT Protocol (below),   Protocolized Standard Care (below)
        • Screen Shot 2014-09-16 at 10.36.07 PM.png             Screen Shot 2014-09-16 at 10.50.41 PM.png
      • Primary outcome: Death from any cause at 60 days.  (Secondary outcomes death at 90 days and 1 year)
        • 60 day mortality was
          • EGDT: 92 deaths (21.0%)
          • Protocol-Based Standard Care: 81 deaths (18.2%)
          • Usual Care: 86 deaths (18.9%) -

               p=0.83 - no significant difference in mortality.

         

        • The two protocol-based approaches led to small but transient improvement in BP by the end of the 6-hour resuscitation time, but higher requirement for ICU and renal replacement therapy.
           
        • NO DIFFERENCE IN 60 AND 90 DAY MORTALITY!
      • Discussion:
        • Mortality rates differ from River's trial. Why?
          • Protocol was slightly changed.
          • Different inclusion criteria.  (Rivers patients were older, had higher initial lactate, higher rates of pre-existing heart and liver disease.
          • (Approximately equal resuscitation fluids given in both trials. - 20-30 ml/kg)
          • Rivers had lower initial ScvO2 readings (49%).
            • Therefore: Rivers et al may have had more severe and persistent shock than in the ProCESS cohort.
              • Counter-argument: subgroup analysis of sicked people shows no difference.
          • Rivers had near perfect adherence compared to (11.9% and 4.4%) in ProCESS but arguable it would have made a difference.  Perfect adherence in practice is hard.
          • Other changes in ICU care changed (lung-protection, tighter glucose control, lower transfusion tresholds)
        • Study lacked the power to identify which groups of patients benefited from protocol-based therapy.

    New Evidence - ARISE and ProMISe Trials (Pending Results)

    • Still in progress as of Sept 2014.

    Adjunctive Therapies

    • Many investigated to reverse persistent hypotension and end organ damage.  
    • Most failed to improve mortality
    • Thrapies with NO BENEFIT:
      • Anti-Infammatories (Ibuprophen)
      • Monoclonal antibody vs. endotoxin
      • Antibody to TNF-alpha and its receptor.
      • IL-1 antagonists
      • Platelet activating factor antagonists
    • Therapies with POTENTIAL BENEFIT
      • Corticosteroids

        • Debate for decades,
        • Many studies show no benefit, and can harm patients (in earlier studies)
          • CORTICUS Trial: Hastens recovery from shock but no effect on survival.  At this time only indicated if suspected adrenal insufficiency, or if unable to get sBP >90mmHg despite aggressive fluid resuscitation and vasopressors.
          • Surviving Sepsis Campaign: Use hydrocortisone only in patients refractory to IV fludis + vasopressor therapy.
        • Also known to help patients with adrenal insufficiency.
        • Usual treatment is ~200-300 mg of IV hydrocortisone (in 2-3 divided doses) daily x7 days
      • Drotrecogin alpha  (aka Protein C)

        • Studies show Protein C levels are low and septic patients are unable to activate it.
        • Protein C inhibits coagulation and inhibits monocyte activation.
        • Some evidence for:
          • In animals: infusion of Protein C reduced mortality in E.coli infections.
          • In humans: modest reduction of mortality in septic shock.
        • This agent is called drotrecogin alpha -- Adjunct to severe sepsis.
        • Reduces mortality to 24.7% from 30.8% placebo control over 27 days.
        • Complications:  BLEEDING
          • Hence only used in intensive care and ID specialists.
          • Major contraindication: recent surgery
      • Glucose Control

        • Glucose control in ICU = reduction in morbidity and mortality
        • Traditionally, treat hyperglycemia if glucose > 215 mg/dL (~11.9 mmo/lL) - ass'd w/ higher risk of infection.
        • "Intensive glucose therapy" 80-110 mg/dL (4.4 - 6.1 mmol/L) initially showed reduction in mortality of post-operative surgical patients.
          • Not reproducible in ICU using same protocol.
        • NICE-SURAR study (Large international study)
          • "Lenient" 180 mg/dL (10 mmol/L) vs. "strict" 81 to 108 mg/dL (4.5 to 6.0 mmol/L)
          • Lenient control = lower mortality 
        • Therefore: Once stabilized, patients with severe sepsis and hyperglycemia in ICU should be receiving IV insulin to target glucose target of <180 mg/dL (<10 mmol/L)
          • NOTE: ACP Practice Guideline: Target 140-200 mg/dL (7.8-11.1 mmol/L)
      • Other
        • Mechanical Ventilation
          • Required if ARDS.
          • NIPPV is not effective, and only delays intubation in septic/ARDS patients.
          • Low tidal volumes should be used to protect lungs.
        • Malnutrition: Avoid!
        • Ventilator Weaning:
          • Therapist-Initiated Weaning Protocols: do not wait for daily physician orders.
          • Daily sedation interruptions
          • Avoid neuromuscular blockade
          • Avoid excessive ventilation
          • Use of daily bolus instead of infusions.
          • (All these improve ventilator weaning)
        • DVT prophylaxis (LMWH/UFH, leg compression devices)
        • Avoid stress ulcer prophylaxis with H2 blockers or PPI
        • IHD or CRRT for sepsis-induced AKI (equivalent, but CRRT may offer easier management)
        • IV Bicarbonate: Do not use to improve hemodynamics or vasopressor requirements in lactic acidosis and pH > 7.15

     

    Surviving Sepsis Guideline Summary 2012

    • Link to guidelines here
    • Recommendation Evidence/Recommendation
      Early quantitative resuscitation of the septic patient during the first 6 h after recognition 1C
      Blood cultures before antibiotic therapy 1C
      Imaging studies performed promptly to confirm a potential source of infection UG

      Broad-spectrum abx therapy within 1 h of the recognition of septic shock

      (and severe sepsis without septic shock)

      1B
      R/A abx daily for de-escalation, when appropriate 1B
      Infection source control (balance of risks and benefits) within 12 h of diagnosis 1C
      Initial fluid resuscitation with crystalloid 1B
      Consideration of the addition of albumin in patients who continue to require substantial amounts of crystalloid to maintain adequate MAP 2C
      Avoid hetastarch formulations  1B
      Initial fluid challenge: (if hypovolemia/hypoperfusion): 30 mL/kg of crystalloid 1C
            Continue fluid challenge if evidence of improvement UG
      Norepinephrine is 1st choice agent to keep MAP > 65 mmHg 1B
      Epinephrine if additional agent need to maintain MAP 2B

      Vasopressin (0.03 U/min) added to Norepi to increase MAP or decrease norepi dose

      (Do not start as first line)

      UG
      Dopamine is not recommended except in highly selected circumstances 2C

      Dobutamine in the presence of

           (a) myocardial dysfunction as suggested by elevated cardiac filling pressures and low cardiac output

           (b) ongoing signs of hypoperfusion despite achieving adequate intravascular volume and adequate MAP

      1C
      AVOID hydrocortisone in adult septic shock if fluid resuscitation and vasopressor therapy were able to restore hemodynamics 2C
      Hemoglobin target 70-90 (7-9g/dL) in absence of tissue hypoperfusion, ischemic CAD or acute hemorrhage. 1B

       

    References / Further Reading

    Tag page (Edit tags)
    • No tags
    Pages that link here
    Page statistics
    39483 view(s), 29 edit(s) and 29138 character(s)

    Comments

    You must login to post a comment.