Slide 1 - ERS Education

Document Sample
Slide 1 - ERS Education Powered By Docstoc
					Thank you for viewing this presentation.
  We would like to remind you that this
material is the property of the author. It is
  provided to you by the ERS for your
 personal use only, as submitted by the

            2008 by the author
Animal Models of Acute
 Lung Injury/ARDS
      Gregory P. Downey M.D.
       National Jewish Health
 University of Colorado, Denver, USA
• Key features of (human) ARDS we trying
    to model
•   Animal models of ALI
     • Direct
     • Indirect
     • Combination
•   Inciting stimulus
     • Endotoxin/LPS
•   Experimental considerations
•   Summary
Key Point 1

• Human acute lung injury (ALI) and
 ARDS have a specific definition with key
 physiological and pathological features
      ARDS: American-European
    Consensus Conference Definition

• Acute onset of respiratory
• Bilateral infiltrates on CXR
• PA wedge pressure < 18 or
  no clinical evidence of LA
• Acute lung injury
   • PaO2/FiO2 < 300
   • PaO2/FiO2 < 200
                                 Bernard et al ARRD 1994
        Acute Lung Injury/ARDS
              Predisposing Factors
Direct Lung Injury               Indirect Lung Injury
Pneumonia                        Sepsis

Aspiration of gastric contents   Trauma/shock

Fat emboli                       Acute pancreatitis

Pulmonary contusion              Multiple transfusions

Inhalational injury              Cardiopulmonary bypass

Reperfusion lung injury

(Induced) Lung Injury
           Physiological Features
• Physiological dysfunction of the lungs
   • Hypoxemic respiratory failure
   •  Lung compliance (stiff lungs)
   • Diffuse (but not uniform) high permeability
       pulmonary edema
•   Evidence of endothelial dysfunction
     •  Circulating levels of endothelial proteins (VWF,
       Factor VIII)
•   Evidence of epithelial dysfunction
     • High protein (albumin) levels in edema fluid
     • Failure of epithelial transport function
        • Inability to clear fluid/solute from the lungs
                Pathological Features
• Diffuse alveolar damage (DAD)
• Key features:
• Neutrophilic alveolitis
   • Evidence of an inflammatory response in the distal lung (alveoli)
• Deposition of hyaline membranes
   • Composed of fibrin and other proteinaceous debris
   • Evidence that serum proteins have entered and precipitated in the
      airspaces  disruption of the alveolar/capillary membrane
• Formation of microthrombi
   • Evidence of endothelial injury and intraluminal activation of the
      coagulation cascade
• Animal model of ALI/ARDS should reflect some of these features
   • Absence of one feature does not mean it is not a form of ALI
Alveolar Edema
ARDS: An Inflammatory Lung Disease

                          Bachofen and Weibel
Pathology of Human ARDS
   Evolving Lung Injury

                    Ware and Matthay NEJM 2000
Key Point 2

• Some but not all features of human
 acute lung injury (ALI) and ARDS
 can be modeled in animals
Animal Models of ALI/ARDS
• Consider which main features of human lung injury
    you are trying to model experimentally
•   Direct or indirect lung injury
•   Inciting stimulus (animal model) = Predisposing
    factor (human ALI/ARDS)
•   Temporal aspects  rapid onset (hrs to days)
•   Endothelial vs epithelial dysfunction
•   Systemic aspects of a model
     • Hypotension, acidosis
     • Dysfunction of other organs may complicate
       interpretation of data
•   The specific characteristics of any given model will
    guide your choice
ALI/ARDS Models: Direct Lung Injury
  • Models in which the lung is injured directly by a
      noxious stimulus
  •   Intratracheal or intranasal administration of
      microbial pathogens or microbial products
       • Bacteria, LPS, LTA, Zymozan
  •   Intratracheal acid (e.g. HCl) ± gastric particulates
       • Reproduce some aspects of aspiration
  •   Hyperoxia
  •   Depletion of surfactant by saline lavage
  •   Ischemia/reperfusion injury
  •   Injurious mechanical ventilation with high tidal
      volumes and low PEEP
       • High stretch (VILI)
ALI/ARDS Models: Indirect Lung Injury

  • Models in which the lung is injured
  • Attempt to reproduce aspects of clinical
      • e.g. sepsis, shock
  •   IV or IP administration of bacteria or LPS
  •   Cecal ligation and puncture (CLP)
  •   Mesenteric ischemia/reperfusion
  •   Transfusion (TRALI)
ALI/ARDS Models: Combinations

  • Combination of different injurious
  • May more accurately reflect human
    disease states
  • „2-hit‟ concept
     • Saline lung lavage + injurious
        mechanical ventilation
    •   CLP + hemorrhagic shock
    •   TRALI + pulmonary infection
Key Point 3

• Different animal models have
 specific strengths and weaknesses
 Pros and Cons of Animal Models
• No single animal model reproduces all of the pathologic
   findings of human ALI/ARDS
     • Even when all 3 elements of the “ALI triad” are present  usually milder
       than in humans ARDS
• Models differ in the impact that they have on the individual
   elements of the ALI triad
    • In some models PMN alveolitis is more marked (e.g., LPS instillation into
       the lungs)
    • In other models the main feature is  intra-alveolar proteinaceous material
       (e.g., ischemia reperfusion of the lungs)
• Sepsis models (e.g. i.v. LPS) usually associated with deposition
   of PMN in pulmonary vasculature but mild  permeability
    • Less intra-alveolar infiltration of PMN or protein deposition.
    • Lung injury in sepsis models localized primarily to the vascular and
       interstitial compartments of the lungs
Which Inciting Stimulus to Use
• Chemicals
     • HCl
     • Oleic Acid
•   Physical forces
     • Injurious mechanical ventilation
•   LPS
     • Well defined, reproducible
     • Single (few) TLR family receptors involved
        • TLR4, TLR2
        • Dependent on purity of LPS
• Live bacteria
   • More physiological but variable
   • Lab strains vs clinical isolates
• Endogenous GI flora
   • CLP
   • Mesenteric ischemia/reperfusion
           Key Features of Animal Models of
                  Direct Lung Injury
    Model               Systemic                  Lung Features
LPS (IT)           Mild hypotension       Acute: large increase in PMN in
                                          airspace with variable changes in
                                          epithelial permeability
                                          Repair phase: Heals without fibrosis
Acid Aspiration Mild decrease in CO       Acute: areas of necrosis, acute
                                          neutrophilic inflammation, hemorrhage
                with hypotension, ileal
                                          and intra-alveolar and interstitial
                permeability              edema,  airway resistance,  lung
                                          compliance, pulmonary HTN
                                          Repair phase: Heals with fibrosis
VILI (high tidal   Variable hypotension   Acute phase: increased pulmonary
volume MV)                                vascular permeability and edema, PMN
                                          infiltration, and sometimes
                                          Repair phase: unclear as model is
                                          usually terminal
           Key Features of Models of
             Indirect Lung Injury

    Model            Systemic                      Lung Features
LPS I.V.        Initial myocardial        Acute: PMN accumulation in capillaries
                depression with           and interstitium but minimal intra-
                systemic hypotension      alveolar migration, mild epithelial
                followed by recovery;
                                          Repair phase: Heals without fibrosis
                Initial pulmonary HTN
LPS I.P.        Myocardial depression, Acute: PMN accumulation in capillaries
                hypotension            with minimal migration into the alveoli,
                                          mild permeability
Live Bacteria   Hypotension, CO          Acute: Interstitial edema, intravascular
I.V.                                      congestion, increased sequestration of
                                          PMN in alveolar capillaries; minimal
                                          epithelial damage, few intra-alveolar
                                          PMN infiltrates, or protein deposition in
                                          the airspaces, pulmonary HTN
           Key Features of Models of
              Indirect Lung Injury

   Model              Systemic                  Lung features
Mesenteric       Hypotension,           Acute: Intra-alveolar
ischemia         metabolic acidosis     proteinaceous exudates and PMN
reperfusion                             sequestration in the vasculature
Cecal ligation   Myocardial depression, Acute: Increased epithelial
and puncture     hypotension            permeability, PMN accumulation in
                                        the interstitium and alveolar
Endotoxin/LPS Models of
   Acute Lung Injury
Endotoxin (Lipopolysaccharide)
Endotoxin/LPS Models of Lung Injury

  • Most biological effects of LPS due to Lipid A
       • Presence of repeating oligosaccharide O
         antigen influences magnitude of response
  •   Mechanism of action of i.v. LPS endothelial
  •   Signaling of LPS is complex
       • Binds to LPS binding protein (LBP)
       • Signals through CD14/TLR4 via MyD88
Host Response to Endotoxin: Toll Receptors

Endotoxin/LPS Models of Lung Injury

  • LPS preparations often contain contaminants,
      such as bacterial lipoproteins
       • May influence the biological effects of LPS
       • TLR4 vs TLR2
  •   LPS can be administered in several ways
       • Systemically  i.v. or i.p.
       • Directly to the lung  aerosol, intranasal,
  •   Optimal dose of LPS depends on the species and
       • Should be determined individually prior to a
Pros and Cons of Animal Models of ALI/ARDS
          Involving Endotoxin/LPS
Model                    Advantages               Disadvantages
LPS I.V.                 Good model of sepsis     Lung injury does not
                                                  mimic human
                                                  Minimal intra-alveolar
                                                  PMN infiltrates and
                                                  protein-rich alveolar
                                                  LPS alone does not
                                                  model all effects of live
LPS I.T.                 Good model of            Fewer changes to the
                         neutrophil recruitment   epithelial barrier
Mesenteric ischemia      Good models of sepsis Injury localized
reperfusion, CLP, I.V.                         primarily to vascular
LPS                                            and interstitial
                                               compartments of lung
Delay in PMN Apoptosis Is Associated
      with  Lung Inflammation
  Empty Vector             Bcl-2 Transduced

      Mice exposed to inhaled LPS (48 hr)
Cecal Ligation and Puncture (CLP)

• Abdominal incision  cecum ligated and
    punctured 3 - 5x with needle
     • Severity of the injury dependent on the number of
      holes in the cecum and on the size of the needle
      used to make the holes
• Resultant peritonitis with systemic
    manifestations of sepsis
•   Applicable to rats and mice
•   Mortality high
    • 25% at 18 hr
    • 70 - 90% in rats by 30 hr
Cecal Ligation and Puncture (CLP)

 • Many similarities to human similar to ALI/ARDS
    • Less intra-alveolar inflammation and hyaline membrane
 • Relatively mild lung injury
 • Probably best animal model of sepsis and multiple
   organ dysfunction
 • Main disadvantage is requirement for major
 • Actual bacterial inoculums in CLP unknown
 • Outcome can be affected by differences in colonic
   flora among animals  significant variability
    Which Species to Choose?
• Advantages of murine models
     • Availability of reagents (e.g. Abs)
     • Access to genetically modified strains  mechanistic
•   Disadvantages
     • Difficulty in doing physiologic measurements  small size
     • Mice differ in some important ways from humans
        • Mice lack IL-8 (key neutrophil chemoattractant in humans)
        • Murine PMN lack -defensins
        • Anatomy of lungs differs from humans
        • Divergence of murine and human TLRs (TLR4 and TLR2)
•   Advantages of rat models
     • Larger size means surgical techniques technically easier
     • Larger blood volume  PMN isolation
•   Disadvantages
     • Unique cytokine/chemokine profiles to be aware of
        • CINC1/2, LIX
     • Transgenics limited
  Which Species to Choose?
• Larger animals
   • Ferrets, pigs, rabbits, dogs, non-human primates (monkeys
     and baboons)
   • Express IL-8
   • Better for complex physiologic measurements
   • Pulmonary intravascular macrophages (PIM) present in
     sheep, calves, pigs, and cats but not dogs or rodents
      • Presence of PIM makes species sensitive to small doses of LPS
• Disadvantages
   •   Expense
   •   Lack of genetically modified strains
   •   Requirement large amounts of reagents
   •   Lack of certain reagents
        • Note utility of human reagents in primate models
• Many experimental models of ALI
• No one model completely reproduces all the features
    of human ALI/ARDS
•   In human disease, PMN (inflammation), hyaline
    membrane deposition (disruption of the
    alveolar/capillary barrier) and microthrombi
    (endothelial injury) are prominent
•   In most experimental models of ALI only one feature
     • Alveolar PMN in LPS instillation
     • Epithelial injury in acid aspiration
•   No single “best” model of lung injury
•   The optimal model is the one that best reproduces
    the features that can be tested by your hypothesis

•   Gus Matute-Bello (Seattle)
•   Tom Martin (Seattle)
•   Michael Matthay (San Francisco)
•   Steve Groshong (Denver)
•   Art Slutsky (Toronto)
    What Parameters to Measure?
• Histological assessment of lung injury
   • Make certain lungs are fixed in inflation
        • 20 cm H2O
     • Semi-quantitative scoring system
        • Proteinaceous debris in alveolar space / hyaline membranes
        • PMN in alveolar space
        • Alveolar wall thickening
•   Evidence of neutrophil infiltration
     • Whole lung MPO
     • BAL neutrophils
     • Number of PMN in lung digests
     • Morphometric analysis
Assessment of Permeability
• BAL total protein is not sufficient and may be
     • Degradation of cells
•   Wet/dry weight
•   Leak of high molecular weight molecules into BAL
     • Endogenous: IgM, albumin
     • Exogenous: 125I-labelled albumin, FITC-dextran
•   Evans blue dye leak
     • Dye binds to albumin
Evidence of Physiological Dysfunction

  • Impairment of gas exchange
     • Blood gases
     • Oximetry
  • Decreased lung compliance
     • Standard PV-curves
     • Flexivent system
Choice of Anesthetics
• Ketamine /xylazine
   • Be cautious of cardiac depressant effects of xylazine
• Ketamine/acepromazine
• Avertin
   • Painful to animals and systemic effects
• Inhaled anesthetics
   • E.g. enflurane
   • Require small animal anesthetic machine for control of
• Potential need for post-op analgesic
   • NSAIDs may modify subsequent inflammatory
    Publishing Your Data
• Report measurements that provide definitive evidence
  that lung injury has occurred in a given model
   • Try to measure at least 2 independent parameters
   • PMN inflammation alone insufficient
• Indirect parameters are not sufficient
   •  cytokines or other molecules
   • Do not provide full evidence of lung injury because these
     changes can occur in the absence of significant
     histopathological changes or altered lung permeability
• Physiologic measurements alone (hypoxemia, change
  in compliance) not adequate indicators of ALI
   • May be caused by conditions such as cardiogenic pulmonary
     edema in which no lung injury has occurred
    Publishing Your Data
• Best evidence of whether ALI has occurred in an
    animal is provided by combination of parameters
•   Measurement of the cellular response
     • e.g. PMN counts in the BAL
•   Measurement of the integrity of the alveolar
    capillary barrier
     • e.g. movement of high molecular weight
       proteins from the serum into the airspaces
•   Histological images showing lung injury plus semi-
    quantitative analysis

Shared By: