2009 03 10 Contrôle de la glycémie et amélioration du ... 03 10 Controle de la glycemie... · Contrôle de la glycémie et amélioration du pronostic des affections graves muscles

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



� Before / after implementation of an IIT protocol

� Clinical outcomes :� Reduction in mortality



� 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


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


Nice-Sugar

Stopped213,500ICU mortality


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



� 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)


Group IIT

(n = 550)

Group LIT

(n = 551) P

Age, yr 65 (51-74) 65 (51 – 74) 0.9207

Median (IQR)


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)


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)


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)


< 0.00012 (0 – 5)0 (0 – 1)Insulin free days, days



P

Group LIT

(n = 551)

Group IIT

(n = 550)


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



P

Group LIT

(n = 551)

Group IIT

(n = 550)


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


SEMIN

AIRE

S IRIS

24

300

250

200

150

100

50

Group IIT


Group LIT

p < 0.0001


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


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


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)


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)


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


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







J Clin Invest 2004; 114;1187J Clin Invest 2004; 114;1187

SEMIN

AIRE

S IRIS

31


MetabolicMetabolic effectseffects

CHOCHO--relatedrelated (relief of glucose (relief of glucose toxicitytoxicity))

CHOCHO--independentindependent


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







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)??







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




INSULIN THERAPYPreiser Devos Crit Care Med 2007;35


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


SEMIN

AIRE

S IRIS

40

0

2

4

6

8

SOFA score

Without Withhypoglycemia

p < 0.01


N = 69N = 1032

0

2

4

6

8

DailySOFA score

DaysWithout Withhypoglycemia

p < 0.01


SEMIN

AIRE

S IRIS

41

Multivariable analysis: hypoglycemia < 60 mg/dl


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


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


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


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


Predictors of brain energy crisis(multivariate logistic regressionadjusted for ICP and CPP) :Serum glucose and dose of insulin


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







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

Documents

2009 03 10 Contrôle de la glycémie et amélioration du ... 03 10 Controle de la glycemie... · Contrôle de la glycémie et amélioration du pronostic des affections graves muscles