brain dysfunction during sepsis - cnerea.fr · concept of sickness behavior • behavior is a...
TRANSCRIPT
BRAIN DYSFUNCTION
DURING SEPSIS
University of Versailles
Raymond Poincaré Teaching Hospital
Garches - France
SEPTIC PATIENTS CAN DIE SEPTIC PATIENTS CAN DIE
FROM BRAIN FAILUREFROM BRAIN FAILURE
A 67 years old COPD womanwho developed a septic shockcomplicated suddenly by acomplicated suddenly by acoma (+ divergent eyes)
Sharshar et al – ICM - 2007
DIFFUSE VASOGENIC EDEMA
(BBB alteration)
SEPTIC PATIENTS CAN DIE SEPTIC PATIENTS CAN DIE
FROM BRAIN FAILUREFROM BRAIN FAILURE
A 72 years old woman withhistory of cirrhosis whodeveloped septic shockcomplicated by seizure and
FLAIRFLAIR
complicated by seizure andcoma without concomitanthypoxemia and hypotension
ADCADC
DIFFUSIONDIFFUSIONDIFFUSE CYTOTOXIC EDEMA
(microcirculatory alteration)
DEFINITIONDEFINITION
Levy et al - Crit Care Med - 2003
SEPSIS
1. Sepsis is an independant risk factors for agitation
(Jaber et al – ICM - 2003)
2. Sepsis is a major cause of delirium (Ely et al –2. Sepsis is a major cause of delirium (Ely et al –
JAMA - 2004) ~ 50%
3. Encephalopathy is occuring in 32 to 60% of septic
patients (Eidelman et al -JAMA -1996)
SEVERITYSEVERITY
GLASGOW COMA SCALE n APACHE II MORTALITY
15 19 17.2 (6.3) 16%15 19 17.2 (6.3) 16%
13-14 15 20.1 (5.4) 20%
9-12 8 23.1 (5.7) 50%
3-8 8 34.4 (7.0) 63%
Eidelman et al - JAMA - 1996
ELECTROENCEPHALOGRAMELECTROENCEPHALOGRAM
1. Excessive theta
2. Predominant delta30
40
50
60
2. Predominant delta
3. Triphasic waves
4. Burst suppresssion
Young et al JCN 1992; Straver et al Neurol Res 1998
0
10
20
30
Normal Theta Delta Triphasic Burst
suppression
NORMAL MILD SEVERE
ELECTROENCEPHALOGRAM
• Sepsis: independent predictor of electrographicseizure (ESZs) or periodic epileptiform discharges(PEDs) : OR: 4.6, 95% CI 1.9–12.7, p = 0.002.(PEDs) : OR: 4.6, 95% CI 1.9–12.7, p = 0.002.
• Sepsis: poor outcome in 90% versus 55% inpatients with and without ESZs or PEDs. adjustedOR 10.4, 95% CI 3.0–50.7, p < 0.001.
Oddo et al Crit Care Med 2008
SENSORY EVOKED POTENTIALSSENSORY EVOKED POTENTIALS
1. Prolonged SEP subcorticaland cortical peak latenciesin 34% and in 84% of septicpatients
2. Cortical Peak latencycorrelated with APACHEIII.
3. SEP peak latency did notdiffer between sedated andnon-sedated patients
Zauner et al - CCM - 2000; Zauner et al - CCM - 2002
PATHOPHYSIOLOGY
Excessive inflammatory brain signalling
(Sickness behavior => Delirium)
MPL (Sharshar 2002)
Neuroinflammatory process CytokinesNeuroinflammatory process (Delirium => neurodegenerative)
Ischemic process
(Delirium/focal => vascular demantia)
(Sharshar 2007, Polito 2011)
(Sharshar 2007, Sharshar 2004)
Cytokines
NO
40% mortality
60% mortality
NEUROINFLAMMATORY
PROCESS
1. Endothelial activation
2. Alteration of blood-brain barrier
1. Role of TNFa
2. Role of complement
EXPERIMENTAL OR CLINICAL DEMONSTRATION
3. Brain expression of cytokines and iNOS
4. Microglial activation
1. Amplification of neuro-inflammatory process
5. Brain cells oxidative stress
6. Brain cell apoptosis
7. Alteration of neurotransmission
MICROGLIAL ACTIVATION
Van Gool et al – Lancet - 2010
Effect of rivastigmine as an adjunct to usual care with
haloperidol on duration of delirium and mortality in critically ill
patients: a multicentre, double-blind, placebo-controlled
randomised trial
Although a sample size of 440 patients was planned,
after inclusion of 104 patients with delirium, the trial
was halted becausewas halted because
1. Mortality in the rivastigmine group (n=12, 22%)
was higher than in the placebo group (n=4, 8%;
p=0·07).
2. Median duration of delirium was longer in the
rivastigmine group (5·0 days, IQR 2·7—14·2) than
in the placebo group (3·0 days, IQR 1·0—9·3;
p=0·06).
Van Eijk et al - Lancet - 2010
METABOLIC DISTURBANCES
• Liver dysfunction
• Kidney dysfunction
• Drugs toxicity• Drugs toxicity
• Drugs withdrawal
• Electrolytic disturbances
• Glycemia
…
GLUCOSE
HYPERGLYCEMIA NORMOGLYCEMIA
(INSULIN)
Microglial activation and apoptosis
(GLUT5)
Astrocytes inactivation
+ -
Polito et al – Crit Care – 2011; Sonneville et al – JCEM - 2012
CEREBRAL INFARCTSn=22 (39%)
Association with clotting disorder
ISCHEMIC PROCESSISCHEMIC PROCESS
Sharshar et al - ICM – 2007; Eischwald et al - Submitted
CYTOTOXIC EDEMA
Bozza et al – JCBFM - 2010
MICROCIRCULATION
Control
Septic
Taccone et al – Crit Care - 2010
CI: cardiac index
FCD: functional capillary density
ISCHEMIA
(100%) HEMORRHAGE
(30%) IDC
(30%)MICRO
ABSCESSES
(10%)
NEUROPATHOLOGY
Sharshar et al - Brain Pathology - 2004
CEREBRAL BLOOD FLOWCEREBRAL BLOOD FLOW
CONTROVERSIAL
Cerebral blood flow, cerebral vasomotor reactivity
and cerebral metabolic rates have been shown to beand cerebral metabolic rates have been shown to be
decreased or preserved in experimental sepsis and in
septic patients.
Crit Care - 2010; Siami et al - Crit Care Clin - 2008
Should we monitor and “optimize” cerebral blood
flow in patients with septic shock?
Pandharipande et al – CC - 2010
TREATMENT
No specific treatment other than control of sepsisNo specific treatment other than control of sepsis
Persistence or reappearance of delirium often indicates Persistence or reappearance of delirium often indicates
that the sepsis is uncontrolledthat the sepsis is uncontrolled
Cerebral effects of accepted treatments of septic shockCerebral effects of accepted treatments of septic shock
1.1. Activated protein C ? (Effect on endothelial activation)Activated protein C ? (Effect on endothelial activation)
2.2. Corticosteroids? (decrease in PTSD, effect on BBB)Corticosteroids? (decrease in PTSD, effect on BBB)
3.3. Insulin? (Insulin? (neuroprotectiveneuroprotective?)?)
MECHANISMS OF LONG-TERM
COGNITIVE DECLINE IN SEPSIS
Psychological
disorders
BEHAVIOR
(AmygadalDELIRIUM
Cognitive
disorders
HIPPOCAMPUS, AMYGDALA, LIMBIC SYSTEM…
CONCEPT OF BRAINSTEM
DYSFUNCTIONDYSFUNCTION
NEUROLOGICAL EXAMINATION
OF SEDATED PATIENTSOF SEDATED PATIENTS
NEUROLOGICAL ASSESSMENT
BRAINSTEM RESPONSESFOCAL SIGNS
Comparison between right
Verbal responseEyes response Motor response
GLASGOW COMA SCALE
1. Eyes spontaneous movement
2. Eyes position
3. Oculocephalogyre response
4. Pupillar size
5. Pupillar light reflex
6. Corneal reflex
7. Grimace
8. Cough reflex
Comparison between right
and left body
1. Motor responses to order
or painful stimulation
2. Limbs tone
3. Tendon reflexes
4. Plantar reflex
Fitting set
n = 72
Validation set
n = 72
Women (%) 24 (33) 28 (39)
Age (years) 58 (46 to 74) 69 (51 to 80)
Surgical admission (%) 16 (22) 22 (31)
CHARACTERISTICS AND OUTCOME
Surgical admission (%) 16 (22) 22 (31)
SAPS-II at admission 50 (37 to 61) 57 (45 to 67)
Duration of sedation (days) 5 (2 to 8) 3 (2 to 6)
Confusion/delirium at awakening (%) 26 (43) 26 (53)
Coma (%) 11 (18) 14 (23)
Altered mental status (%) 34 (57)* 34 (55)**
Mortality rate at day 28 (%) 22 (31) 21 (29)
* 60 and ** 62 patients
NEUROLOGICAL EXAMINATION
12-24H OF SEDATION Fitting set Validation set
Number of patients 72 72
Midazolam (mg/kg) 0.9 (0.6 to 1.8) 1.3 (0.8 to 2.0)
Subfentanyl (µg/kg) 2.0. (0.8 to 4.0) 2.0 (0.7 to 4.6)
sedation to inclusion (hours) 12 (12-24) 12 (12-24)
ATICE (from 0 to 20) 9 (9 to 10) 9 (9 to 10)
RASS Not tested -4 (-4 to -2)RASS Not tested -4 (-4 to -2)
Blinking to strong light (%) 31 (43) 28 (39)
Absent eye movement (%) 66 (93) 67 (93)
Myosis (%) 45 (63) 38 (54)
Pupillary light response (%) 51 (71) 58 (82)
Corneal reflex (%) 65 (90) 66 (92)
Oculocephalic response (%) 32 (47) 33 (46)
Cough response (%) 36 (51) 60 (83)
Grimacing (%) 41 (57) 48 (69)
Myo
91
0
Cluster Dendrogram
Crn: corneal reflex
Blk: blinking
Gmg: grimace
Cgh: cough
NEUROLOGICAL EXAMINATION
SYSTEMATIZED ALTERATION OF BRAINSTEM REFLEXES BY SEDATION
Gm
g
Cg
h
Ocr B
lk
Myo
Eyp
Crn Lid
Lig
ht
56
78
agnes (*, "ward")
t(x)
Heig
ht
Cgh: cough
Light: pupillar reflex
Moi: miosis
EyP: eye position
Ocr: oculocephalogyre
Septic shock with ARDS and
severe liver and renal failure
in a aplasic 82 years old man.
1. Coma plus abolition of all
brainstem reflexes
DISCREPANCY
brainstem reflexes
2. After discontinutaion of
sedation, recovery of
cough and oculocephalic
repsonses but not of
corneal reflex and
grimace.
ALTERED MENTAL STATUS
(after discontinuation of sedation)
Fitting set Validation set
RESPONSES ASSESSED BETWEEN THE 12Th AND 24th H OF SEDATION
Multiple logistic model
Fitting set Validation set
Criteria Confusion or coma Delirium or coma
OR (95%CI) P OR (95%CI) P
SAPS-II at inclusion 1.04 (1.00 to 1.07) 0.058 1.03 (0.99 to 1.08) 0.10
Absent oculocephalic response 4.49 (1.34 to 15.1) 0.015 5.64 (1.63 to 19.5) 0.006
Sharshar et al – Crit Care Med - 2011
Multiple logistic model
28-DAYS MORTALITY
Fitting set Validation set
RESPONSES ASSESSED BETWEEN THE 12Th AND 24th H OF SEDATION
Sharshar et al – Crit Care Med - 2011
Fitting set Validation set
OR (95%CI) P OR (95%CI) P
SAPS-II at inclusion 1.06 (1.02 to 1.09) 0.003 1.03 (1.00 to 1.07) 0.051
Absent cough response 7.80 (2.00 to 30.4) 0.003 5.44 (1.35 to 22.0) 0.017
C-index (SE) 0.836 (0.055) 0.743 (0.067)
CONCEPT OF BRAINSTEM
DYSFUNCTION
Oculocephalic response
Apoptosis LC
RAAS dysfunctionDELIRIUM
CRITICAL
ILLNESS
Apoptosis LC
Effect of DEX
DEATHAutonomic dysfunction
Cough reflex
HR/BP variability
Immune control
IMMUNO
DEPRESSION
Inappropriate Sympathetic Activation at Onset of
Septic Shock: A spectral Analysis approach
Annane et al - AJRCCM - 1999
NEURO-IMMUNE MODULATION
ΣΣΣΣ PARAΣΣΣΣ
αααα2 ββββ2ββββ1 αΑαΑαΑαΑ7
(CMganglion)
DYSFUNCTION
HPA axis
Splenic AR
INFLAMMATION
+ _
EXPERIMENTAL SEPSIS-RVLM
LPS IV
oligodendrocytes
Superoxide (O2-)
iNOS
neurons
astrocytes
microglia
Apoptosis
T1
Cytochrome C
Apoptosis
Arterial hypotension
Chan et al-Free Rad-2005
T2 T3
T4
PROJECT
Clinical and neurophysiological assessment ofClinical and neurophysiological assessment of
relationships between brainstem response and
delirium and death in critically ill patients with and
without brain damages
SEDATION
(C’est la confusion!)
BZD
(Dose/durée)
DXM Interruption
quotidienne
DELIRIUM0
-+
Physiothérapie
-
Absence
sédation
+
Confort/amnésie ?
Faux souvenirs
BE CAREFULBE CAREFUL
In a 65 years old man lightlysedated for a hypoxemicpneumonia:
+ no response of the leftarm to painful stimuli
+ a greater hypotonia of theleft arm and leg
Facial Facial
palsypalsy
left arm and leg
+ a decreased contraction ofthe right face to painfulstimulation
+ Claude-Bernard Hornersign
FOCAL NEUROLOGICAL SIGNS IN A CRITICALLY ILL PATIENT
WITH CEREBROVASCULAR RISK FACTORS AND TREATED BY HEPARIN
WeaknessWeakness
BE CAREFULBE CAREFUL
Accounts for all neurological signs
BE CAREFULBE CAREFUL
Three days after discontinuation of sedation, the patient
developed agitation and delirium. General and neurological
examination was not changed. EEG was normal. Agitation and
delirium was ascribed to discontinuation of sedation. Three days
later, patient complained of pain of the legs and ankles that
turned to be due to bacterial and ischemic myositis.turned to be due to bacterial and ischemic myositis.
DELIRIUM WITHOUT (new) FOCAL NEUROLOGICAL SIGN
IN A RECENTLY NON SEDATED PATIENT
DIFFERENTIATING
ENCEPHALOATHY FROM ENCEPHALOATHY FROM
SICKNESS BEHAVIOR
CONCEPT OF SICKNESS BEHAVIOR
• Behavior is a major, highly preserved and adaptivecomponent of response to stress.
• The behavioral response to stress is variable, ranging fromaggressiveness, anxiety or hyperalertness to lethargy and isaggressiveness, anxiety or hyperalertness to lethargy and ismainly controlled by the amygdala and hippocampus,which are vulnerable to hemodynamic and metabolicinsults.
• Changes in behavior are therefore not easily interpretablebecause they can be adaptive or maladaptive, physiologicor pathophysiologic.
Dantzer et al – Nat Neurosci Rev
PHYSIOLOGICAL BRAIN SIGNALLINGSickness behavior
Adapted neurovegetetative and neuroendocine response
Vagal nerve
Dantzer et al – Nat Rev Sci - 2008
Circumventricular organs
(no BBB)
Vagal nerve
Delirium+
NE NE
BHV
(Amygadal DELIRIUM
IMMUNE
Psychological
disorders
NE
NV
(Brainstem
NE
COMA
IMMUNE
SYSTEMCognitive
disorders
CONCLUSION
SEPSIS ASSOCIATED ENCEPHALOPATHY IS
1. A FREQUENT AND SEVERE COMPLICATION
2. HAS TO BE DETECTED ROUTINELY
3. HIS ASSESSMENT IS MAINLY CLINICAL, EEG OR MRI BEING
UNDERTAKEN UPON NEUROLOGICAL ARGUMENT
4. REQUIRES A SYSTEMATIC APPROACH THAT TAKE INTO ACCOUNT
MENGITIS, UNCONTROLLED SEPSIS AND DRUG NEUROTOXICITY
5. SECONDARY TO ISCHEMIC AND INFLAMMATORY PROCESSES,5. SECONDARY TO ISCHEMIC AND INFLAMMATORY PROCESSES,
INVOLVING BRAIN PERFUSION, MICROCIRCULATION AND BBB AND
MICROGLIAL CELLS
6. ASSOCIATED WITH LONG TERM COGNITIVE DYSFUNTION
ITS TREATMENT IS:
1. CONTROL OF SEPSIS,
2. AVOID NEUROTOXICITY,
3. NEUROPROTECTIVE AGENTS OR STRATEGY?
Raymond Poincaré
Service de Réanimation
MERCI
Institut Pasteur
Fabrice Chrétien
Histopathologie Humaine et Modèles Animaux
ACKNOWLEDGEMENTS
Institut PasteurRaymond Poincaré
Dr Romain Sonneville
Dr Nicolas Weiss
Dr Andréa Polito
DrShidasp Siami
Dr Juliette Bailly-Salin
Dr Anne-Laure le Maho
Beaujon
Dr Jeremy Allary
Dr Stanilas candlemanDr Nicolas Weiss
Pr Fabrice Chretin
Pr Mervyn Singer
Dr Anne-Laure le Maho
Dr Aurelien Mazeraud
Dr Eric Azabou
Pr Geoffroy Lorin
Pr Djillali Annane
Pr Tarek Sharshar
Dr Stanilas candleman
Pr Jean Mantz
UCL
Pr Greet van den Berghe
Leuven
Jan-Pieter Konsman
Bordeaux
Non sedated
EXAMINATION Normal
LUMBAR PUNCTURE IF ANY DOUBT
Sedated
Discontinuation of sedation
What
strategy?
Sedation necessary
Monitoring
Focal sign
Delirium
Agitation
Coma
Myoclonus
Brain imaging
EEG - Biochemistry - Drugs Brain imaging
EEG - Biochemistry - Drugs Brain imaging
Biochemistry - Drugs Brain imagingEEG
UNCONTROLLED SEPSIS?
Antagonist?
ME
CH
AN
IMS
OF
SE
PS
IS A
SS
OC
IAT
ED
EN
CE
PH
AL
OP
AT
HY
ME
CH
AN
IMS
OF
SE
PS
IS A
SS
OC
IAT
ED
EN
CE
PH
AL
OP
AT
HY
Raymond Poincaré
Service de Réanimation
MERCI
Institut Pasteur
Fabrice Chrétien
Histopathologie Humaine et Modèles Animaux