2009 03 10 contrôle de la glycémie et amélioration du ... 03 10 controle de la glycemie... ·...
TRANSCRIPT
1
Jean-Charles Preiser
Séminaire des services d’urgences IRISBruxelles, 10 mars 2009
Diabète et urgences
Contrôle de la glycémie et amélioration
du pronostic des affections graves
muscles
glucoseglucose
cerveau
acides grasacides gras
érythrocytes
LymphocytesG.Blancs
intestin
glutamine
foiefoie
TissusTissusagressagress ééss
lactatelactate
Adaptations métaboliques à l’agression
Insulino-résistance����
alaninealanine
adipocytes
glycéro ll
Insulino-résistance����
InsulinoInsulinoindind éépendancependance
SEMIN
AIRE
S IRIS
2
TIGHT GLUCOSE CONTROL:
The dream comes true
BenefitsBenefits
••EasilyEasily accessibleaccessible
••CheapCheap
••ReducesReduces complication ratecomplication rate
••ReducesReduces mortalitymortality
••DecreasesDecreases LOSLOS
SEMIN
AIRE
S IRIS
3
TIGHT GLUCOSE CONTROL:
The dream comes true
BenefitsBenefits
••EasilyEasily accessibleaccessible
••CheapCheap
••ReducesReduces complication ratecomplication rate
••ReducesReduces mortalitymortality
••DecreasesDecreases LOSLOS
RisksRisks -- constraintsconstraints
••HypoglycemiaHypoglycemia
••EquippmentEquippment
••HumanHuman resourcesresources
� Monocentric, retrospective� 1600 patients (medico-surgical) � Glycemia target : <140 mg/dl
� Mortality 14.8% Vs 20.9% (p <0.01)
EFFECTS OF AN INTENSIVE MANAGEMENT
PROTOCOL ON MORTALITYKrinsley Mayo Clin Proc 2004;79:992
SEMIN
AIRE
S IRIS
4
GLUCOSE CONTROL AND MORTALITY IN
CRITICALLY ILL PATIENTSFinney JAMA 2003;290:2041
� 530 patients : cardiothoracic surgery (90%)� Ranges of glycemia :
� 0.8-1.1� 1.1-1.4� 1.4-1.8� 1.8-2.0� >2.0
� Decreased mortality when glycemia < 1.4 g/l� « Control of glucose levels rather than of absolute levels of
exogenous insulin appear to account for the mortality benefit withintensive insulin therapy »
DIGAMI 1 studyMalmberg JACC 1995;26:55 - BMJ 1997; 314:1512
� 620 patients with diabetes and AMI randomized to� Glucose-insulin for > 24 hours
� Standard treatment
� Outcome :� Reduction in HbA1c� Sustained reduction in mortality (3-y)
SEMIN
AIRE
S IRIS
5
PORTLAND DIABETIC PROJECTFurnary Endocr Pract 2004;10(S2):21
� 3,896 patients with diabetes undergoingcardiac surgery procedures
� Insulin IV vs s/c
� RRR of death 57%� RRR of deep sternal wound infection 66%
INTENSIVE INSULIN PROTOCOL IN
TRAUMA PATIENTSReed J Am Coll Surg 2007;204:1048
� Before / after implementation of an IIT protocolSE
MINAIRE
S IRIS
6
INTENSIVE INSULIN PROTOCOL IN
TRAUMA PATIENTSReed J Am Coll Surg 2007;204:1048
� Before / after implementation of an IIT protocol
� Clinical outcomes :� Reduction in mortality
INTENSIVE INSULIN PROTOCOL IN
TRAUMA PATIENTSReed J Am Coll Surg 2007;204:1048
� Before / after implementation of an IIT protocol
� Clinical outcomes :� Reduction in mortality� Reduction in the frequency of intraabdominal
abscesses (2.7 % to 0.7 %, p<.01)� Reduction in ventilator days (3.1 + 0.3 to 2.4 +
0.2, p < .05)
SEMIN
AIRE
S IRIS
7
� All patients admitted to a surgical ICU (n = 1548)� Continuous intravenous insulin to control blood glucose� Randomized to 2 different blood glucose targets� Intensive treatment � 4.4 – 6.1 mmol/L
versus� Conventional treatment � 10.0 – 11.1 mmol/L� 200 – 300 g iv glucose on first day, EN or TPN thereafter
Greet Van den Berghe, M.D., Ph.D., P. Wouters, M.Sc., F. Weekers, M.D., Ch. Verwaest, M.D.,
Intensive Insulin Therapy in Critically Ill Patients
N Engl J Med 2001; 345 1359
Intensive insulin therapy : Mortality
Result Control Intensive %. p
1. ICU mortality (%) 8.0 4.6 - 47% < 0.004
� First 5 d. of ICU stay (%) 1.8 1.7 NS
� ICU stay > 5d (%) 20.2 10.6 - 48% 0.005
� Diabetic pat. > 5d (%) 20.6 10.7 - 48% 0.005
2. Hospital mortality (%) 10.9 7.2 - 34% 0.01
Intensive treatment ���� 4.4 – 6.1 mmol/L versusConventional treatment ���� 10.0 – 11.1 mmol/L
N Engl J Med 2001; 345 1359
SEMIN
AIRE
S IRIS
8
Squares : Squares : glycemiaglycemia < 110 mg/dl< 110 mg/dl
CirclesCircles: : glycemiaglycemia 110110--150 mg/dl150 mg/dl
Triangles: Triangles: glycemiaglycemia > 150 mg/dl> 150 mg/dl
CUMULATIVE RISK OF DEATH IN ICU CUMULATIVE RISK OF DEATH IN ICU PATIENTS PATIENTS
Van den Berghe Van den Berghe CritCrit Care Med 2003;31:359Care Med 2003;31:359
RRR = Relative RRR = Relative riskrisk reductionreduction
NNT = NNT = NumberNumber neededneeded to to treattreat
SECONDARY OUTCOME VARIABLES
SEMIN
AIRE
S IRIS
9
In-hospital survival in the intention-to-treat analysis and in the subgroup of patients with ICU stay > 3 days
-7% (p=0.31)
-18%(0.05)
No effect on survival by intention-to-treat analysis,Beneficial effect in patients with ICU stay > 3 days
VN Engl J Med 2006; 345: 449
DearDearDearDear Greet,Greet,Greet,Greet,
I I I I agreeagreeagreeagree withwithwithwith youyouyouyou
on the on the on the on the toxictoxictoxictoxic effectseffectseffectseffects of of of of severesevereseveresevere hyperglycemiahyperglycemiahyperglycemiahyperglycemia
on the on the on the on the benefitbenefitbenefitbenefit of Intensive of Intensive of Intensive of Intensive InsulinInsulinInsulinInsulin therapytherapytherapytherapy in in in in LeuvenLeuvenLeuvenLeuven
I I I I feelfeelfeelfeel concernedconcernedconcernedconcerned by by by by
the case mixthe case mixthe case mixthe case mix
the the the the amountamountamountamount of IV glucoseof IV glucoseof IV glucoseof IV glucose
the proportion of patients the proportion of patients the proportion of patients the proportion of patients receivingreceivingreceivingreceiving steroidssteroidssteroidssteroids
the the the the delaydelaydelaydelay beforebeforebeforebefore improvementimprovementimprovementimprovement
the the the the risksrisksrisksrisks of of of of hypoglycemiahypoglycemiahypoglycemiahypoglycemia
the the the the workloadworkloadworkloadworkload
the issue of the issue of the issue of the issue of bloodbloodbloodblood glucose glucose glucose glucose variabilityvariabilityvariabilityvariability
the absence of the absence of the absence of the absence of benefitsbenefitsbenefitsbenefits in in in in somesomesomesome categoriescategoriescategoriescategories of patients (of patients (of patients (of patients (diabeticsdiabeticsdiabeticsdiabetics LOS <3d)LOS <3d)LOS <3d)LOS <3d)
Greet Van den BergheUZ Leuven
Greet Van den BergheUZ Leuven
SEMIN
AIRE
S IRIS
10
CALORIC INTAKE
101364673840SomeEN
979677758785Predom
PN
143 +60
141 +62
179 +64
179 +65
161 +64
160 +66
IV glucose
10 + 510 + 519 + 720 + 615 + 815 + 8Calories
IITCITIITCITIITCIT
<3d<3d>3d> 3dITTITT
Van den Berghe et al Van den Berghe et al DiabetesDiabetes 2006;55:31512006;55:3151
AmtAmt insulininsulin : 70 : 70 ±± 2 U/d 2 U/d vsvs 12 12 ±± 1 U/d1 U/d
Daily Daily caloriccaloric intakeintake : 8 to 24 kcal/: 8 to 24 kcal/kd.dkd.d
InsulinInsulin/calorie ratio/calorie ratio
Van den Berghe et alVan den Berghe et alCritCrit Care Med 2003; 31:359Care Med 2003; 31:359
INTAKE OF CALORIES AND INSULIN
SEMIN
AIRE
S IRIS
11
0
20
40
60
80
100
120
140
160
180
200
Mean amount of
glucose/day (g)
Without TPN With TPN
LITIIT
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
12
SEMIN
AIRE
S IRIS
13
RESTORING
« NORMOGLYCEMIA » IMPROVES SURVIVAL !
YES� Observational findings
� DIGAMI 1� Furnary� Reed� Krinsley� Finney
� Interventional data� Leuven 1 study
RESTORING
« NORMOGLYCEMIA » IMPROVES SURVIVAL !
YES� Observational findings
� DIGAMI 1� Furnary� Reed� Krinsley� Finney
� Interventional data� Leuven 1 study
NO� Observational findings
� DIGAMI 2� CREATE-ECLA� Treggiari
� Interventional data� Leuven 2 study *� Gandhi� VISEP � GLUCONTROL
SEMIN
AIRE
S IRIS
14
Intensive insulin therapy and mortality in critically ill
patientsMiriam M Treggiari, Veena Karir, N David Yanez, Noel S Weiss, Stephen Daniel
and Steven A Deem
Critical Care 2008, 12:R29 (doi:10.1186/cc6807)
125
130
135
140
145
150
Period I Period II Period III
Mean BG
0
2
4
6
8
10
12
Period I Period II Period III
ICU mortality
Cohort study comparing three consecutive time periods – total 10,456 patients :
- period I no protocol (n = 2,366 03/01- 02/02)
- period II target BG 80-130 mg/dl (n= 3,322, 03/02-06/03 ),
- period III target BG 80-110 mg/dl (n= 4,786 , 07/03-02/05)
INTENSIVE INTRAOPERATIVE INSULIN
THERAPY DURING CARDIAC SURGERYGandhi Ann Intern Med 2007;146:233
� 400 Adults undergoingelective cardiacsurgery randomisedto intraoperativeinsulin therapy(target 80-100 mg/dl)vs conventionaltreatment
SEMIN
AIRE
S IRIS
15
INTENSIVE INTRAOPERATIVE INSULIN
THERAPY DURING CARDIAC SURGERYGandhi Ann Intern Med 2007;146:233
� Results� Composite primary outcome end-point (30-day death,
sternal infection, prolonged ventilation, cardiacarrhythmias, stroke and renal failure) unchanged (44 vs 46%)
� More deaths (4 vs 0) and strokes (8 vs 1, p < .05) in the IIT group
CURRENT MULTI-CENTRE STUDIES ON
TIGHT GLUCOSE CONTROL IN ICUS
236,10090-d mortality
Open labelRandom/ctrlStratified
Nice-Sugar
193,500ICU mortality
Open labelRandom/ctrlStratified
Glucontrol
17600Mortality2x2
Randomfluid + BG
VISEP
Current
status
Nb hosp
Nb pts
required
Prim
End-pt
DesignSEMIN
AIRE
S IRIS
16
CURRENT MULTI-CENTRE STUDIES ON
TIGHT GLUCOSE CONTROL IN ICUS
Recruitmentstoppedmid-Aug2008
236,10090-d mortality
Open labelRandom/ctrlStratified
Nice-Sugar
Stopped213,500ICU mortality
Open labelRandom/ctrlStratified
Glucontrol
Stopped17600Mortality2x2
Randomfluid + BG
VISEP
Currentstatus
Nb hosp
Nb pts
planned
Prim
End-pt
Design
TARGETS FOR BLOOD GLUCOSE
140-18080-110NICE-SUGAR
140-18080-110GLUCONTROL
180-20080-110VISEP
180-20080-110Leuven
Target in LITTarget in IITStudySEMIN
AIRE
S IRIS
17
VISEP STUDY
488 patients in 17 centres
28-d mort 90-d mort Rate hypo
IITCIT
35.4%35.4% 36.7%36.7%
0%0%
25%25%
50%50%
75%75%
24.7%24.7%26.0%26.0%
0%0%
25%25%
50%50%
75%75%
2828--day mortalityday mortality 9090--day mortalityday mortality
Mor
talit
y (%
)M
orta
lity
(%)
n=288n=288 n=247n=247 n=288n=288 n=247n=247
Intensive Insulin Therapy (SepNet)- Mortality in severe sepsis -
p=0.73p=0.73 p=0.33p=0.33
Conventional InsulinConventional Insulin Intensive InsulinIntensive Insulin
SEMIN
AIRE
S IRIS
18
0 10 20 30 40 50 60 70 80 90 100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Days
< 110 mg/dl
110 -150 mg/dl
p= 0.99 (log -rank test)
Ove
rall
Sur
viva
l
106 95 90 84 79 78 77 74 74 0
138 127 114 107 101 95 94 94 93 0
Patients at Risk :
< 110 mg/dl:
110 -150 mg/dl120
152
166 154 132 124 121 117 113 112 112 0> 150 mg/dl183
> 150 mg/dl
0 10 20 30 40 50 60 70 80 90 100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 10 20 30 40 50 60 70 80 90 100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Days
< 110 mg/dl
110 -150 mg/dl
p= 0.99 (log -rank test)
Ove
rall
Sur
viva
l
106 95 90 84 79 78 77 74 74 0
138 127 114 107 101 95 94 94 93 0
Patients at Risk :
< 110 mg/dl:
110 -150 mg/dl120
152
166 154 132 124 121 117 113 112 112 0> 150 mg/dl183
> 150 mg/dl
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
� Prospective, randomised, controlled, investigator-blinded and multicentric study
� Aimed at comparing the effects of two regimens of insulin therapy, respectively titrated to achieve a blood sugar level � between 4.4 and 6.1 mmol/l (80 - 110 mg/dl) : GROUP IIT� and between 7.8 and 10.0 mmol/l (140 - 180 mg/dl) :
GROUP LIT
SEMIN
AIRE
S IRIS
19
� Inclusion criteria� 18 years or older
� admitted in an ICU
� Exclusion criterion� Absence of signed informed consent
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
� 7 countries� Austria, Belgium, France, Israel, The
Netherlands, Slovenia and Spain.
� 21 units in 19 centres
SEMIN
AIRE
S IRIS
20
� 1108 recruited patients� 1101 randomized patients� Length of observation:
� from 1 to 56 days (median: 5; IQR: 3 – 10)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Group IIT
(n = 550)
Group LIT
(n = 551) P
Age, yr 65 (51-74) 65 (51 – 74) 0.9207
Median (IQR)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
21
Group IIT
(n = 550)
Group LIT
(n = 551) P
Age, yr 65 (51-74) 65 (51 – 74) 0.9207
Sex ratio, M/F 352/198 338/213 0.3827
Category
Medical
Scheduled Surgery
Emergency Surgery
Trauma
42.9 %
31.3 %
18.1 %
7.7 %
41.2 %
32.7 %
18.1 %
7.9 %
0.9437
Median (IQR)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Group IIT
(n = 550)
Group LIT
(n = 551) P
APACHE II score 15 (11 - 21) 15 (11 – 21) 0.797
SOFA score 7 (5 – 9) 7 (5 – 9) 0.467
GCS 15 (9 -15) 15 (9 – 15) 0.708
Median (IQR)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
22
<0.000157.784.6Patients treated by insulin IV (ITT), %
<0.000166.996.8Patients treated by insulin IV (PP), %
P
Group LIT
(n = 551)
Group IIT
(n = 550)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
< 0.00012 (0 – 5)0 (0 – 1)Insulin free days, days
<0.000157.784.6Patients treated by insulin IV (ITT), %
<0.000166.996.8Patients treated by insulin IV (PP), %
P
Group LIT
(n = 551)
Group IIT
(n = 550)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Duration of insulintreatment, hrs
96 (40-213) 87 (41-227) 0.6588
SEMIN
AIRE
S IRIS
23
< 0.00012 (0 – 5)0 (0 – 1)Insulin free days, days
< 0.00010.4 (0.0 – 1.4)1.8 (1.0 –2.9)
Insulin rate, U/hr
<0.000157.784.6Patients treated by insulin IV (ITT), %
<0.000166.996.8Patients treated by insulin IV (PP), %
P
Group LIT
(n = 551)
Group IIT
(n = 550)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Duration of insulintreatment, hrs
96 (40-213) 87 (41-227) 0.6588
Changes of insulinrate, number per day
3 (2-5) 1(0-3) 0.0001
80
90
100
110
120
130
140
150
160
170
180
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Treatment, days
Blood glucose, mg/dl
IIT
LIT
Median with IQR
* * * * * * * * **
* * * * *
* p < 0.001
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
24
300
250
200
150
100
50
Group IIT
Blood glucose, mg/dl
Group LIT
p < 0.0001
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Number of BG checks/patients:Number of BG checks/patients:
From 2 to 856 measures (median: 33; IQR: 14 From 2 to 856 measures (median: 33; IQR: 14 -- 85)85)
250
200
150
100
50
Blood glucose, mg/dl
p < 0.001For eachcomparison
TARGET IS NOT
ALWAYS REACHED
Target range (CIT)
Target range (IIT)
112
151
VISEPVISEP LeuvenLeuven II LeuvenLeuven IIII GlucontrolGlucontrol
153
103
153
111
131
106
SEMIN
AIRE
S IRIS
25
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
0
2
4
6
8
10
12
14
16
18
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Group IITGroup LIT
Number of inclusions
Cum
ulative ICU deathrate, %
0 30 60 90 120 150 180 210 240 270 300 330 360
Time, days
100
90
80
70
60
50
40
30
20
10
0Survivalprobability(%
)
Group LIT
Group IIT
Logrank test: p = 0.7412
Hazard ratio: 0.949 (95 % CI: 0.695 - 1.296)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
26
Group IIT
(n = 550)
Group LIT
(n = 551) P
ICU death rate, % 16.7 15.2 0.5022
Hospital death rate, % 24.5 20.7 0.1452
Day 28 death rate, % 19.5 16.2 0.1685
Median (IQR)
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Univariable analysis
Crude OR 95 % CI p
Group IIT 1.14 0.82 - 1.14 0.427
Multivariable analysis
Adjusted OR 95 % CI p
Group IIT
Gender (male)
Age, yr
Apache II
SOFA
1.24
1.32
1.01
1.12
1.03
0.84 – 1.85
0.87 - 1.32
0.99 – 1.02
1.09 – 1.16
0.95 – 1.11
0.276
0.196
0.133
< 0.0001
0.459
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
RISK OF DEATHRISK OF DEATHSEMIN
AIRE
S IRIS
27
EFFECTS OF IIT ON MORTALITYProspective randomised multi-centre controlled trialsMerz Finfer Crit Care 2008; 12:212
EFFECTS OF IIT ON MORTALITYProspective randomised multi-centre controlled trialsMerz Finfer Crit Care 2008; 12:212
SEMIN
AIRE
S IRIS
28
THE AVAILABLE META-ANALYSES
0.94 (0.8-1.11)0.95 (0.85-1.03)
0.69 (0.51-0.99)
0.85 (0.75-0.97)
RR (95% CI) for mortality (TGC vsstandard care)
NNYYGIK included
ICU ICU SurgeryCardiac surgeryType of patients
5403(530 (10-1548))
8432(89(10-1548))
5150(34 (14-1548))
8478 (72(14-1548))
Number of patients
8293435Number of studiesincluded
30611358445941Number of studiesretrieved
1966-2002 1948-2008Years coveredGriesdaleWienerGandhiPittasFirst author
1966-20081976-2006
RiskRisk of of deathdeath
«« The The availableavailable mortalitymortalitydata data representsrepresents onlyonly 40%40%of the optimal informationof the optimal informationsize size requiredrequired to to reliablyreliablydetectdetect a plausible a plausible treatmenttreatmenteffecteffect. Possible. Possiblemethodologicalmethodological and and reportingreportingbiasesbiases weakenweaken inferencesinferences. . »»
PERIOPERATIVE INSULIN INFUSIONGandhi Mayo Clin Proc 2008
SEMIN
AIRE
S IRIS
29
Benefits and Risks of Tight Glucose Control
in Critically Ill AdultsWiener Wiener Larson JAMA 2008;300:933
� 29 trials – 8432 ICU patients � Hospital mortality did not differ between TGC and usual care
(21.6 vs 23.3%, RR 0.93 (0.85-1.03))
� No significant difference in mortality after stratification� By glucose goal (< 110 vs < 150 mg/dl)� By type of units (medical surgical or mixed)
� No significant decrease in need for new dialysis
� Decreased risk of septicemia (10.9 vs 13.4 % (RR 0.76 (0.56-0.97), significant only in the moderately TGC (target < 150 mg/dl)
� Increased risk of hypoglycemia (13.7 vs 2.5 (RR 5.13 (4.09-6.43))
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;v35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?
� Associated workload
SEMIN
AIRE
S IRIS
30
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;v35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?
� Associated workload
J Clin Invest 2004; 114;1187J Clin Invest 2004; 114;1187
SEMIN
AIRE
S IRIS
31
J Clin Invest 2004; 114;1187J Clin Invest 2004; 114;1187
MetabolicMetabolic effectseffects
CHOCHO--relatedrelated (relief of glucose (relief of glucose toxicitytoxicity))
CHOCHO--independentindependent
J Clin Invest 2004; 114;1187J Clin Invest 2004; 114;1187
MetabolicMetabolic effectseffects NonNon--metabolicmetabolic effectseffects
CHOCHO--relatedrelated (relief of glucose (relief of glucose toxicitytoxicity))
CHOCHO--independentindependent
SEMIN
AIRE
S IRIS
32
PUTATIVE METABOLIC
MECHANISMS OF INSULIN
� Relief of glucose toxicity� Upregulation of Glut-4� Mitochondrial damage� Enhanced oxstress� Enhanced NO� PMN function
� CHO-independent effects� Reversal of hypertriglyceridemia� Increase HDL and LDL-
cholesterol� Decrease FFA� Increased muscle protein
content (decreased breakdown or increased production?)
PUTATIVE NON-METABOLIC
MECHANISMS OF INSULIN
� Attenuation of inflammatory reaction (CRP, MBL), pro/anti-inflammatory cytokines
� Anti-apoptotic effects� Prevention of endothelial dysfunction� Prevention of hypercoagulability and platelet
hyperaggregability� Reduction in ADMA levels� Reduction of tissular edema� Reduction of collagen / immunoglobulin glycosylation
SEMIN
AIRE
S IRIS
33
INSULIN IS MUCH MORE THAN A
GLUCOSE-LOWERING HORMONE !
Adapted from Bouglé – Annane 2009
Effects of IIT on CRP in children/neonates
Vlasselaers Lancet 2009 doi:10.1016/S0140-6736(09)60044-1
SEMIN
AIRE
S IRIS
34
Copyright ©2008 The Society of Thoracic Surgeons
Albacker T. et al.; Ann Thorac Surg 2008;86:20-27
Intraoperative blood glucose and insulin levels
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;v35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?
� Associated workload
SEMIN
AIRE
S IRIS
35
Minimal and maximal tolerated
blood glucose levelsBelgian Survey 2007 (Preiser Sottiaux Crit Care 2008)
0%
10%
20%
30%
40%
50%
60%
40 50 60 80 100 120 140 150 160 180 200
Blood Glycemic Level
Minimum tolerated Maximum tolerated
WHICH IS THE MEANING OF
« NORMOGLYCEMIA » IN THE ICU?
� 80-110 mg/dlis considered asNormoglycemia infasting conditions
� Stress� Feeding� Therapies
CommentaryRestoring normoglycaemia: not soharmlessJean-Charles PreiserPublished: 28 February 2008 Critical Care 2008, 12:116 (doi:10.1186/cc6787)
SEMIN
AIRE
S IRIS
36
Optimal target for blood glucose
� In Leuven : 80 - 110 mg/dl (4.4-6.1 mmol/L)� Elsewhere :
� Below 180 mg/dl (10.0 mmol/L)� Higher than 80 mg/dl (4.4 mmol/l)� Below 150 mg/dl (7.8 mmol/L)??
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;v35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?
� Associated workload
SEMIN
AIRE
S IRIS
37
JAMA 2006 ; 295 : 1681JAMA 2006 ; 295 : 1681SEMIN
AIRE
S IRIS
38
160
140
120
100
80
60
40
20
0
Variability of blood glucose
Group IIT Group LIT
SD of blood
glucose, mg/dl
p NS
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?� Associated workload
SEMIN
AIRE
S IRIS
39
0
5
10
15
20
25
30
Leuven I Leuven II VISEP Glucontrol Arabi De la Rosa
Control Intensive
Incidence of hypoglycemia during IIT Prospective studies
Increased risk of hypoglycemia (13.7 vs 2.5 (RR 5.13 (4.09-6.43))
0
10
20
30
40
ICU Mortality (%)
Without Withhypoglycemia
N = 69
N = 1032
p < 0.001
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
40
0
2
4
6
8
SOFA score
Without Withhypoglycemia
p < 0.01
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
N = 69N = 1032
0
2
4
6
8
DailySOFA score
DaysWithout Withhypoglycemia
p < 0.01
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
SEMIN
AIRE
S IRIS
41
Multivariable analysis: hypoglycemia < 60 mg/dl
Adjusted OR 95 % CI p
Group IIT
Death
Apache II
7.05
2.19
1.07
4.72 - 10.53
1.38 – 3.48
1.04 – 1.10
< 0.0001
0.0008
< 0.0001
Multivariable analysis: hypoglycemia < 40 mg/dl
Adjusted OR 95 % CI p
Group IIT
Death
Apache II
4.29
2.26
1.07
2.10 – 8.76
1.15 – 2.26
1.03 – 1.11
0.0001
0.0177
0.0008
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
RISK FACTORS FOR
HYPOGLYCEMIAVriesendorp et al Crit Care Med 2006; 34:96
83126Insulin
217Lower nutrition
114CVVH bicar
826CVVH
1431Sepsis
2047Diabetes
5275SOFA shock>1
5275Female
No hypo (n = 155)Hypo (n = 156)
RetrospectiveRetrospective collection ; collection ; hypoglycemiahypoglycemia < 45 mg/dl< 45 mg/dl
SEMIN
AIRE
S IRIS
42
SEVERE HYPOGLYCEMIA : RISK FACTORS
AND OUTCOMEKrinsley Grover Crit Care Med 2007;35:2262
� 102 patients with at least one episode of severe hypoglycemia (< 40 mg/dl) matchedwith 306 control patients from a cohort of 5,365 patients
SEVERE HYPOGLYCEMIA : RISK FACTORS
AND OUTCOMEKrinsley Grover Crit Care Med 2007;35:2262
� Mortality 55.9 % in patients with severehypoglycemia vs 39.5 in non-hypoglycemicpatients (p < .01)
� Multivariable logistic regression analysisidentified hypoglycemia as an independentrisk predictor of mortality (OR 2.3[1.4-3.7])
SEMIN
AIRE
S IRIS
43
0
0,5
1
1,5
2
2,5
3
3,5
Leuven I Leuven II VISEP Glucontrol Arabi
Relative risk of death (vs non-hypoglycemic)
Relative risk of death of patients with
hypoglycemiaProspective studies
DURATION OF HYPOGLYCEMIA
DURING IIT
0
40
80
120
160
200
Time (min)
Duration of hypoglycemia < 40
IIT LIT
P<0.001
GLUCONTROLGLUCONTROLGLUCONTROLGLUCONTROL
Brain interstitial glucoseDecreased by IIT(Vespa et al Crit Care Med 2006)
SEMIN
AIRE
S IRIS
44
POTENTIAL RISK FACTORS
FOR HYPOGLYCEMIA
� Patient-related :� General condition� Variations in insulin
sensitivity� Adrenal failure� Liver failure� Renal failure
� Treatment-related� Enteral nutrition
(tolerance, interruption)� Insulin algorithm IV
glucose� Substitution fluid for
CVVH
Potential toxicity of hypoglycemia
� Hypoglycemia-associated autonomic failure(HAAF)?
� Neurologic� Related to duration?
� Non-neurologic� Delayed diagnosis of adrenal / liver failure (falsely
attributed to IIT)� Others ?
SEMIN
AIRE
S IRIS
45
Hypoglycemia and the brain
� Glucose is the obligatory metabolic fuel for the injured brain
� No cerebral stores of glucose� Glucose diffusion from plasma to
neurons and astrocytes (concentration-dependent)
� In case of severe hypoglycemia, fall of ATP and cortical activity (EEG)
� Potential roles of lactate / glycogenreleased from astrocytes as rescuesubstrates ?
Impact of TGC on cerebral glucose metabolismOddo et al Crit Care Med 2008;36:3233
SEMIN
AIRE
S IRIS
46
Impact of TGC on cerebral glucose metabolismOddo et al Crit Care Med 2008;36:3233
Predictors of brain energy crisis(multivariate logistic regressionadjusted for ICP and CPP) :Serum glucose and dose of insulin
Impact of TGC on cerebral glucose metabolismOddo et al Crit Care Med 2008;36:3233
Predictors of hospital mortality(logistic regression)Brain energy crisis 7.4 (1.4-39.5)*Glasgow Coma scale 1.1 (.96-1.3)CPP 1.01 (.97-1.04)ICP 1 (0.99-1.01)
SEMIN
AIRE
S IRIS
47
IS TGC SAFE?
� Risk of hypoglycemia increased 4- to 6- foldwhen applying tight glucose control (BG target 80-110 mg/dl)
� Incidence of hypoglycemia increased in mostseverely ill patients
� Causal relationship between hypoglycemiaand in mortality – neurological damage not completely excluded
Meanwhile :
Moving beyond tight glucose control to safe
effective glucose controlJames S Krinsley and Jean-Charles Preiser
Critical Care 2008, 12:3: 149
Instead of TGC, we propose a stepwise approach defining anew standard – Safe, Effective Glycemic Control (SEGC).SEGC involves, first, adoption of a safe glycemictarget appropriate to the skills, experience and available toolsof the ICU that does not result in a significant increase in therate of hypoglycemia. A glycemic target of 80 to 150 mg/dl isnot unreasonable for an ICU to choose initially; implementation can subsequently lead to downward revisionof the glycemic goal.
SEMIN
AIRE
S IRIS
48
CLINICAL EXPERIENCE WITH TIGHT
GLUCOSE CONTROL BY INTENSIVE
INSULIN THERAPYPreiser Devos Crit Care Med 2007;v35
� How does IIT work?� Optimal target for blood glucose� Is the absolute level or the variability of blood
glucose the most detrimental factor?� Is hypoglycemia life-threatening?
� Associated workload
0 10 20 30 40 50 60 70
Other
No perceived obstacles
Concern about the accuracy of POC glucose meters toguide glycemic control care
Lack of financial resources
Reluctance to have patients endure the pain of frequentfingersticks
Lack of familiarity with hospital inpatient glycemic controlguidelines
Not convinced of the benefits of glycemic control
Lack of nursing endorsement
Lack of local expertise in inpatient hyperglycemia/diabetesmanagement
Lack of standardized institutional policies related toglucose management
Lack of administrative resources
Lack of clinical resources
Lack of physician endorsement
Concern about causing hypoglycemia
Percentage of Respondents
Cook BC et al SCCM congress (poster #282)
Perceived obstacles to the implementation of TGC
US survey
SEMIN
AIRE
S IRIS
49
0
2
4
6
8
10
12
14
16
18
Time to reachtarget (h)
Maintenancein target(h/day)
Incidencesevere hypo
(%)
InterventionControl
2 2 cohortscohorts of 50 patientsof 50 patients
Control : Control : physicianphysician’’ss discretiondiscretion
Intervention : Intervention : StandardizedStandardized protocolprotocol((targettarget 4.5 4.5 –– 6.1 6.1 mmolmmol))
Number BG checks (mean +/- SD)
0
20
40
60
80
100
120
140
160
180
200
IIT LIT
p < 0.001
Number BG checksPerreaux et al Intensive Care Med 2007
SEMIN
AIRE
S IRIS
50
Number of changes of insulin ratePerreaux et al Intensive Care Med 2007
Changes insulin rate (Median - IQR)
0
1
2
3
4
5
6
IIT LIT
p < .0001
The future? Intravascular continuous
blood monitoring?
SEMIN
AIRE
S IRIS
51
TIGHT GLUCOSE CONTROL WITH
INTENSIVE INSULIN THERAPY
Hazards ofhyperglycemia Risks of
hypoglycemia
Being funambulist may not be accessible to everyone
SEMIN
AIRE
S IRIS
52
BELGIUM IS OPEN-MINDEDThe same story can be read in different ways!
SEMIN
AIRE
S IRIS