20 ans de réanima-on: le choc sep-que - jivd - aer · 20 ans de réanima-on: le choc sep-que. en...
TRANSCRIPT
Didier Payen, MD, Ph D
Lariboisière Hospital – Université Paris 7 UMR INSERM 1160
20 ans de Réanima-on: le choc sep-que
En 20 ans, que s’est-‐il passé? • Sepsis defini-on more or less stable • Epidemiology è
– incidence, age, comorbidi-eskk – Big data èè large cohort
• BeBer educa-on, faster diagnosis, reasonable recomenda-ons (SSC x 3)
• Huge technological progresses: genotyping; transcriptomic; proteomic; metabolomic…
• More clear aspects of immuno-‐inflammatory processes • Reasonable suppor-ve therapy: Fluid amount; pressors; ven-la-on; RRT; ECorp Circ…
En 20 ans, qu’en ai-‐je -ré? • Over simplis-c view leads to mistakes: è RCTs always failed despite solid basic science background
• Infec-on does not kill by itself but by host response – Concept of sep-c phases – Inflamma-on BM; organ damage BM, etc…
• Time for infec-on treatment is crucial (SSC x 3) – Golden hours; AB administra-on; fluid is necessary but not too much! Pressors YES but for what BP level?
• Huge technological progresses: – genotyping; transcriptomic; proteomic; metabolomic… – More rapid detec<on of pathogens
• Elderly pa-ents SHOULD be treated. • Exp models are not easily transposable to human beings
SEVERE SEPSIS
• Sepsis + hypotension (÷÷APSyst > 20 mmHg or APsyst <80 mmHg) or symptoms of hypo-‐perfusion or organ failure :
• AND at least ONE among following criteria
-‐ hypotension (systAP< 90 or ÷÷> 40)
-‐ lac-c acidosis
-‐ oliguria
-‐ encephalopathy
-‐ hypoxemia
-‐ coagulopathy
The W-shaped mortality curve seen in the 1918 influenza pandemic curves for incidence of clinical illness vs. fatality rate markedly differ Same age-dependent mortality pattern is seen for TB, pneumococcal infection, trauma, Yellow fever, malaria, streptococcal Toxic shock, etc. èè What accounts for this difference?
Fedson D. An-viral Research 2013; 99,(3), 417-‐ 435
Sepsis Puzzle
outcome An- infl cytokines
Physiol Status
Type of infec-on
circulatory & metabolic consequences
Iatrogenic complica-ons
Economical aspects
Gene-c Time course of the Disease
Pro infl cytokines
Strategy for care
Hormonal paBerns
Sepsis: we beBer understand and conceptualize …
From the last 20 years, concept have moved on a lot… especially in immuno-‐
inflammatory processes
PAMPS èè mu-ple; PRR DAMPS èè mul-ple; DAMP rec -‐ Necro<c cells à DAMPs -‐ Criteria for DAMPs
-‐ Alone è biological response without contamina<on (PAMPs = 0)
-‐ Being efficient at physiological [ ] -‐ Their blockade èè inhibates their ac-on -‐ Examples: HMGB1; Calgranulines…
From Matzinger theory
• PAMP: Proteins from pathogens are processed in the lysosomes to generate an-genic pep-des, which form a complex with MHC class II on the surface of APC, recognized by T cell receptors (TCR)
APC = ANTIGEN PRESENTING CELLS
• LPS complex has 3 components: CD14, TLR4, and MD-‐2. • TLR4 interacts with CD14. Protein MD-‐2 is required for TLR4-‐mediated recogni-on of LPS • TLR4 and MD-‐2 are cons-tu-vely associated with each other. (CD14 is presumably recruited aNer binding LPS) • Limited informa-on about allelic variants of human toll genes and impact on suscep-bility to infec-on
The signalisation
From Cell death to inflamma-on and OF - Necrotic cell death è DAMPs è receptors
è+ prod of inflam cytokines (IL‑1). - Other molecules proteases; hydrolases act on EC components è + mediators (complement fragments) or DAMPs è prod of inflam cytokines by host cells.
- Pro-inflam mediators è local vascular endothelium lesions è ‘leaky’, attracts neutrophils and monocytes/macrophages è soluble (antibody) and cellular defences in the tissue (cell infiltration) neutralize or contain pathogens
Medzhitov et al. Science 2012; 355:936-‐41
• ““Threat”” vs. non-threat
• Tolerating pathogens
• Limiting host response
• Stress hormones • Anti-inflammatory
actions • Pro-resolving
actions
Organ-‐specific tolerance capacity to pathogen-‐ or immune-‐induced pathology
Medzhitov et al. Science 2012; 355:936-‐41
Organs largely differ in term of : -‐ Suscep-bility -‐ Repair capabili-es -‐ Func-onal autonomy -‐ sequela
Mortality
-‐ Is early mortality (1st Wk) similar to late mortality?
-‐ Is inflamma-on moving along -me? -‐ What is sepsis-‐induced mortality?
-‐ Crude mortality? -‐ ACributable mortality?
0 5 10 15 20 25 30
0 0.
1 0.
2 0.
3 0.
4 0.
5 0.
6
0 5 10 15 20 25 30
0 0.
1 0.
2 0.
3 0.
4 0.
5 0.
6
0 5 10 15 20 25 30
0.1
0.2
0.3
0.4
Low IL -‐ 10/ Low comorbidi-es (n=67) High IL -‐ 10/ Low comorbidi<es (n=71) Low IL -‐ 10/High comorbidi-es (n=42) High IL -‐ 10/High comorbidi<es (n=38)
Pro
babi
lity o
f deat
h
Days Days
Pro
babi
lity o
f deat
h Low IL -‐ 6/ Low comorbidi-es (n=64) High IL -‐ 6/ Low comorbidi<es (n=75) Low IL -‐ 6/High comorbidi-es (n=46) High IL -‐ 6/High comorbidi<es (n=34)
IL-10 IL-6
202 SS Patients (multicentric)
Early death (<7days) relates mainly to inflammation intensity, but not to comorbidity
level but
Late death (>7days), relates mainly comorbidity
This concept obtained from a training cohort (202 pts) was tested in large testing cohort N = 989 SS patients from Prowess data base
N = 989 pa-ents in sep-c shock
0 5 10 15 20 25
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Days
Prob
abili
ty o
f dea
th
Low Il6/Low comorbidities n=167 Low IL6/High comorbidities n=297 High Il6/Low comorbidities n=215 High IL6/High comorbidities n=253
P value < 0.001
Theore<cal crude mortality of 40% in the absence of therapy. Since matched controls had a crude mortality of 20%, è the es<mate of aBributable mortality is 20% (40% -‐20%). è The best Bmt can reduce mortality about max 20% but cannot reduce the mortality due to the underlying diseases.
ATTRIBUTABLE MORTALITY 20 %
The host response to sepsis
Days
Pro
-in
flam
mat
ory
re
s po
nse
s A
nti
-in
flam
mat
ory
re
s po
nse
s Excessive inflammation causing collateral damage (tissue injury)
Immune suppression resulting in enhanced susceptibility for nosocomial pathogens
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 0.4
0.5
0.6
0.7
0.8
0.9
1.0
Survival
Study Day
Log-‐rank test P value: 0.5789
Hazard ra-o (95% CI): 1.05 (0.88, 1.26)
Placebo
Eritoran
Placebo Censored
Eritoran Censored
ACCESS trial: TLR4/MD2 inhibition in severe sepsis
28-day mortality
28.1%
26.9%
PROWESS-SHOCK (NEJM 2012)
Ranieri et al NEJM 2012
Percent in study hospital 28d: DrotAA 305 (49%) Placebo 279 (43%)
-‐ fast track process (short windows: hrs…)
-‐ mul- interac-ve processes
-‐ lack of validated BM to rapidly characterize host
reponse
Early phase is challenging because of:
Why not being inspired by cancer strategies?
cecal liga<on and puncture (CLP) mouse model: 80% of C57BL/6 mice succumbed within 48 hr Ø epirubicin i.p. at the time of CLP and again 24 hr later reproducibly and significantly (p < 0.001) increased the survival Ø Independently from mouse strain Ø “ “ “ “ of sepsis of origins in addition to peritoneal sepsis
Ø epirubicin-treated + CLP has similar bacterial load in blood and organs 24 hr post-CLP
Ø a substantial reduction in the levels of inflammatory mediators TNF, IL-1b, IL-6, and HMGB1
Ø Epirubicin protection against Sepsis is mediated by ATM, a master regulator of the DDR (DNA damage response)
Ø ATM-deficient(Atm-/-) mice were not
protected by epirubicin against CLP and died with similar kinetics to wild-type (WT, Atm+/+) animals
Epirubicin has a 24 hr Therapeu-c Window to Protect against Sepsis
The Host Response to Sepsis
Urgent need for biomarkers that reflect the predominant type of the host response to sepsis in an individual patient. ü Aid in identification of patients that might
benefit from a specific intervention. ü Aid in titrating the intervention and
monitoring the response. ü Rapid: “point-of-care”.
Pathophysiology: Kidney is perfused by inflamed blood: èè activated Immune
Cells + elevated mediators plasma levels èè infiltration
Systemic inflamma-on èè AKI via vascular compartment
q Loss of delayed type hypersensi-vity response to common recall an<gens
q Apoptosis-‐induced deple-on of immune effector cells, loss of CD4, CD8, B, and dendri-c cells q Reac-va-on of latent viruses (CMV; herpes virus in roughly 25–35% of pa<ents with sepsis)
q autopsy èè most pa-ents admiBed to ICUs for treatment of sepsis had unresolved sep-c foci at post mortem, èèpa-ents unable to eradicate invading pathogens and were more suscep-ble to nosocomial organisms, or both q Blood studies from pa<ents with and without sepsis show decreased produc-on of proinflammatory cytokines, decreased monocyte HLA-‐DR expression, increased numbers of regulatory T cells (Treg Fox P3), increased produc-on of PD-‐1 or PD-‐L1
Clinical or laboratory evidence for sepsis being an immunosuppressive disorder
q splenocytes from sepsis pa<ents had highly
significant func-onal impairments as evidenced by major reduc-ons in cytokine secre-on.
q Mul-ple inhibitory mechanisms were iden-fied, ü dominance of inhibitory over ac-va-ng receptors,
ü expansion of suppressive cell types,
ü and induc-on of inhibitory ligands on both APCs and
<ssue parenchymal cells.
q Both pro-‐ and an--‐inflammatory cytokines were
impaired.
q Cytokine produc-on at 22 hours in sepsis pa-ents was comparable
with controls, è some pa-ents, defec-ve cytokine secre-on
may be reversible if cells are removed from the sepsis milieu.
Sepsis èè Profound Loss of CD4 and CD8 Cells the loss of CD 4 T cells mimics HIV in severity èè
in ineffec-ve adap-ve immunity
Non-‐ sep-c Sep-c
Boomer, Chang, Takasu, et. al Journal of the American Med. Associa;on (JAMA) 306:2594 2011
1
10
100
1000
10000
100000
BLOOD BM
Mon
ocyt
e H
LA-D
R (A
B/C
)
p<0.0001 p=0.0009
A ICU Patients; Non ICU patients; HV
Blood and BM mHLA-‐DR expression in pa-ents vs controls
1
10
100
1000
10000
100000
BLOOD BM
Mon
ocyt
e H
LA-D
R (A
B/C
)
B Sshock, Septic, non septic; non ICU, HV
p<0.0001
p=0.0006
p<0.0001 p=0.0008
p=0.0412
p=0.0056
PlosOne (In press)
Conclusion
The downregula-on of HLA-‐DR in BM monocyte
lineage is present in ICU pa-ents without major
changes in myeloid cells. It may result from a
regula-on mediated by soluble and/or neuro-‐
endocrine factors present in BM cell
microenvironment.
59 2013
PMN, Mono phagocytose mHLA-‐DR TNFαα
Fct Mono Phagocytose
Prolif. Tcell IFNγγ An--‐apoptose êêPD-‐1, ãã molec adhesion êêTreg Diversité récepteurs TCR
TotoR Pseudomonas aeruginosa septic shock – severe ARDS – peritonitis in a 15 months post-LT child
0
5000
10000
15000
20000
25000
J6 J8 J11 J14 J15 J18 J26
mHLA-DR
Leuco
Lympho
Neutro
Severe PAIDsn
Normal
Grey zone
PAIDs
AB/C or 10e6
INFγγ
Norepinephrine
Mechanical ventilation Positive cultures Negative cultures
To conclude • PAIDs is a fact that concerns all acute situations, particularly severe sepsis is
always present after 1 or 2 days post injury
• IMMUNOSCOPE – Blood measurements: WBC (fractions), NCLRatio; semi-quantitative cytokine level
– Flowcytometry: HLA-DR; Ly sub-populations
• è IMMUNOSCOPE! èè Personalized therapy
• HLA-DR is a good candidate and is cheap and fast to be measured with
semi-quantitative IL-6, IL-10.
• Immuno-stimulation can be proposed on solid criteria
• RCT are on going testing different targets and molecules (INFγγ, GM-CSF; IL-7…)
interferon γγ
0
5000
10000
15000
20000
25000
30000
35000
Day3 Day6 Day7
Day9 Day11
Day12 Day13
Day15 Day17 discharge
BAL pseudomonas aeru 109 +++ +++ + 0 0
2603 570
mHLA-‐DR
(AB/C)
Days awer ICU admission (post cardiac arrest)
Lukaszewicz et al.Crit Care Med 2009; 37: 2746–2752
Nl
Risk factors for severe sepsis or sep-c shock
Brun-‐Buisson et al, JAMA 1995
• Ageing • Male sex (63 %) • Emergency admission: medical or surgical • Admission in a large hospital: > 400 beds • immunodepression or underlying pathologies with a short term bad prognosis.
EXPRESSION OF ADHERENCE MOLECULES ON ENDOTHELIAL CELL CULTURE èè TOO LONG
SERA WHOLE
BLOOD
SURFACE ANTIGEN EXPRESSION (FLOW
CYTOMETRY)
2hrs
CYTOKINE SECRETION Quick semi-quantitative method IMMEDIATE
WHOLE
BLOOD
PATIENTS
Post-agressive Immuno-Depression Syndrome: PAIDs STRATEGY FOR IMMUNOSCOPING!
Blood as « tissue Biopsy » - WBC - Fractional count - PMN/Ly ratio