Time Dependence of Reperfusion Therapy: “Wavefront” Phenomenon of Necrosis in Dogs

© © CM Gibson 2006CM Gibson 2006

Cross-sections of left ventricle after experimentalcoronary artery occlusion

Reimer KA, et al. Circulation. 1977;56:786-794.

Time Dependence of Reperfusion Therapy:Time Dependence of Reperfusion Therapy:“Wavefront” Phenomenon of Necrosis in Dogs“Wavefront” Phenomenon of Necrosis in Dogs

Duration of occlusion 3 h

Area supplied byoccluded artery

x xx

xx

x

xxx x

xxx

x xx

x

xx x x

xx

xx x xx

xx

x

XXXXNecrosis Ischemic but viable Nonischemic

24 h40 min


Time is Muscle: Delays in Symptom to Door Time are Associated Time is Muscle: Delays in Symptom to Door Time are Associated with Poorer Flow in the Muscle Independent of Flow in Arterywith Poorer Flow in the Muscle Independent of Flow in Artery

16% rise in risk of a closed muscle for every hour of delay in STEMI (p=0.0005) Gibson, JACC 2004

Normal LargeCapillary Lumen

Clinical Picture: Clinical Picture:

Open ArteryOpen Artery

Closed MuscleClosed Muscle

Blistering of CapillariesBlistering of Capillaries

& Endothelium& Endothelium

“Blistered” / Swollen Endothelium /wall of the capillary

Normal non-infarctedwall without edema

Infarcted wall with edema (white areas)

Interstitial Edema Interstitial Edema EHJ (2004) 25, 794–803EHJ (2004) 25, 794–803


0

1

2

3

GUSTO I90-93

2.8

Med

ian

Tim

e (h

rs) B

etw

een

Sym

ptom

O

nset

and

Tre

atm

ent

GUSTO III95-97

InTIME II97-99

ASSENT 300-01

ASSENT II97-98

GUSTO V99-01

Time from Symptom Onset to Study Treatment in Large Scale Thrombolytic Trials

2.8 2.7 2.72.7 2.9

Gibson CM, Circulation 2001;104:2632-2634


Adapted from Tiefenbrunn AJ, Sobel BE. Circulation. 1992;85:2311-2315.

Time-Dependent Benefit of Reperfusion TherapyTime-Dependent Benefit of Reperfusion Therapy

0

20

40

60

80

100

0 2 4 6 8 10 12

Reperfusion Time (hours)

% B

enef

it

Reimer/Jennings 1977Bergmann 1982GISSI-I 1986

© © CM Gibson 2006CM Gibson 2006Adapted from Timm TC, et al. Circulation. 1991;84(suppl II):II-230.

6.2

3.23.7

5.2

01234567

1 2 3 4

Time From Symptom Onset (hours)

Mor

talit

y R

ate

(%)

PP=0.05=0.05

TIMI 2: 6-Week MortalityTIMI 2: 6-Week Mortality


Adapted from Lee KL, et al. Circulation. 1995;91:1659-1668.

Importance of Time-to-Treatment: Results of GUSTO-IImportance of Time-to-Treatment: Results of GUSTO-I

0

2

4

6

8

10

12

0 1 2 3 4 5 6 7 8 9 10 11 12

Time From Onset of Symptoms to Treatment (hours)

2=149 (1 df )

30-D

ay M

orta

lity

( %)


Rapid Treatment Is Associated Rapid Treatment Is Associated With Decreased MortalityWith Decreased Mortality

Data source: Boersma E, et al. Lancet. 1996;348:771-775.

Number of lives saved per 1,000 patients treated with fibrinolytics (based on 35-day mortality)

65

37

2629

0

10

20

30

40

50

60

70

0-1 h 1-2 h 2-3 h 3-6 h

Live

s sa

ved

(No.

)

Time from symptom onset to thrombolysis

50,246 pt meta-analysis1983-93

© © CM Gibson 2006CM Gibson 2006Adapted from Weaver WD, et al. JAMA. 1993;270:1211-1216.

1.2

8.7

4.9

11.2

0.0

2.0

4.0

6.0

8.0

10.0

12.0

Mortality Infarct Size

Tx <70 minutesfrom onset

Tx >70 minutesfrom onset

PP=0.04=0.04 PP<0.001<0.001

Perc

ent

MITI: Mortality, Infarct Size, and TimeMITI: Mortality, Infarct Size, and Time


1.0

3.7 4.06.4

14.1

0

5

10

15

20

25

60 61-75 76-90 91

Berger PB, et al. Circulation. 1999;100:14-20.

PP=0.001=0.001

Door-to-Balloon Time (minutes)

Importance of Door-to-Balloon Time: Importance of Door-to-Balloon Time: 30-Day Mortality in the GUSTO-IIb Cohort30-Day Mortality in the GUSTO-IIb Cohort

Mor

talit

y (%

)

>< PTCA not performed


11.14 1.15

1.411.62 1.61

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0-60

61-9

0

91-1

20

121-

150

151-

180

>180MV

Adju

sted

Odd

s of

Dea

th

Cannon CP, Gibson CM et al, JAMA 2000


Patency and Mode of ReperfusionPatency and Mode of Reperfusion

0

20

40

60

80

100

0 30 45 60 75 90 120 150

Pate

ncy

Rat

e (%

)

Fibrinolysis

PTCA at 75minutes

PTCA at 120minutes

Time (minutes)

90-minutepatency

ED

arrival

Drug

administration

Adapted from Gibson CM. Ann Intern Med. 1999;130:841-847.


Hypothetical Relationship Between Early Reperfusion, Hypothetical Relationship Between Early Reperfusion, Mortality Reduction, and Extent of Myocardial Salvage Mortality Reduction, and Extent of Myocardial Salvage

Gersh BJ, et al. Gersh BJ, et al. JAMAJAMA. 2005;293:979-986.. 2005;293:979-986.

00

2020

4040

6060

8080

100100

1212 2424Time From Symptom Onset to Reperfusion TherapyTime From Symptom Onset to Reperfusion Therapy

(hours)(hours)

Mor

talit

y Re

duct

ion,

(%

)M

orta

lity

Redu

ctio

n, (

%) Mortality Reduction (%)Mortality Reduction (%)

Extent of SalvageExtent of Salvage(% of area at risk)(% of area at risk)

Critical Time-dependent Critical Time-dependent PeriodPeriodGoal: Myocardial SalvageGoal: Myocardial Salvage

D-B – HarmD-B – Harm

A-B – No BenefitA-B – No Benefit

Shifts in Shifts in Potential Potential OutcomesOutcomes

A-C – BenefitA-C – BenefitB-C – BenefitB-C – Benefit

D-C – HarmD-C – Harm

00

Time-independent PeriodTime-independent PeriodGoal: Open Infarct-related Goal: Open Infarct-related ArteryArtery

44

DD

CC

BB AA

88 1616 2020


Door to Balloon Time

<60 min(44 min ± 12)

n=183

60-90 min(75 min ± 9)

n=296

90-120 min (103 min ± 8)n=304

>120min(156 min ± 30)

n=403p-value

Onset-to-Door (mins) 197 ± 182 140 ± 140 141 ± 151 113 ± 124 <0.0001

Onset-to-Balloon (mins) 235 ± 180 216 ± 140 244 ± 150 269 ± 126 <0.0001

Death 30-days 0.6% 0.7% 4.7% 2.5% 0.0037

MACE 30-days 1.6% 2.4% 7.6% 5.5% 0.0034

Infarct Size (% LV) 14.2 ± 15.8 13.2 ± 14.6 18.0 ± 18.0 17.4 ± 18.0 0.0023

Does a 90-Minute Door-to-Balloon Time Matter? Observations From Four Current Does a 90-Minute Door-to-Balloon Time Matter? Observations From Four Current Reperfusion TrialsReperfusion Trials

EMERALD, COOL MI, AMIHOT, ICE-ITEMERALD, COOL MI, AMIHOT, ICE-IT

O’Neill, et al. JACC,2005:Suppl A:45:225A.


Time from Symptom Onset to TreatmentTime from Symptom Onset to TreatmentPredicts 1-year Mortality after Primary PCIPredicts 1-year Mortality after Primary PCI

De Luca et al, Circulation 2004

The relative risk of 1-year mortality increases by7.5% for each 30-minute delay

n=1791


Time from Symptom Onset to TreatmentTime from Symptom Onset to TreatmentPredicts One-year MortalityPredicts One-year Mortality

4.4

1.5

5.74.7

1.2

6.3

8.5

0.8

11.9

9.7

0.0

13.0

0

3

6

9

12

15 p = 0.006

<2 hrs 2-4 hrs 4-6 hrs

p = 0.02

De Luca at al, JACC 2003

>6 hrs

All Patients Low-Risk

p = NS

High-Risk


Time from Symptom Onset to Treatment Predicts One-year Mortality Time from Symptom Onset to Treatment Predicts One-year Mortality Among Patients with Occluded Epicardial ArteriesAmong Patients with Occluded Epicardial Arteries

5.34.1

2.7

5.55.9

9.2

0

12.7

0

3

6

9

12

15 p = 0.013

<2 hrs 2-4 hrs 4-6 hrs

p = NS

De Luca at al, JACC 2003

>6 hrs

Pre-PCI TIMI 2/3 Flow Pre-PCI TIMI 0/1 Flow


Predictors of Door-to-Balloon Delay in Primary Predictors of Door-to-Balloon Delay in Primary AngioplastyAngioplasty

3.7

3.2

2.7

2.2

1.7

Mea

n D

o or -

to-b

allo

on t i

me

(hou

rs)

Mea

n D

o or -

to-b

allo

on t i

me

(hou

rs)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 230 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Time of hospital arrival (24 hour clock)Time of hospital arrival (24 hour clock)

n=40,017

Angeja BG, et al. Angeja BG, et al. Am J CardiolAm J Cardiol 2002;89:1156-1161) 2002;89:1156-1161)


What is Primary Angioplasty ?What is Primary Angioplasty ?

• Primary Angioplasty: Door to balloon time <1 hour

• Secondary Angioplasty: Door to balloon time 1 - 2 hours

• Tertiary Angioplasty: Door to balloon time 2 - 3 hours

• Post MI Care: Door to balloon time 3 - 4 hours

• Elective Case: > 4 hours

CM Gibson 2006


PCI-Related Time Delay vs Mortality Benefitin 22 Randomized Studies of PCI vs Fibrinolytic Therapy

-5

0

5

10

15

Abso

lute

Ris

k Di

ffere

nce

In D

eath

(%)

0 20 40 60 80 100PCI-Related Time Delay (min)

Nallamothu and Bates, AJC 2003

23 RCTs For every 10 min delay to PCI: 1 % reduction in Mortality Difference

Between PCI & LysisN= 7419

p=0.006

DANAMI: on site PCI90 DB – 50 DN = 40 min delay

DANAMI: with transfer110 DB – 50 DN = 60 min delay

“USA AMI” with transfer:171 DB – 32 DN =139 min delay


NRMI 2,3,4

452,544 Reperfusion Eligible

STEMI Patients

1,963 Hospitals

Transfer Out Patients

n=119,235

Missing Time Intervalsn=13,137

Did Not Receive PCIor Fibrinolytic Therapyas Initial Reperfusion

n=89,524

Study Population192,509 Patients

645 Hospitals230,648 Patients1,860 Hospitals

≥20 STEMI Patients Treated

Treatment of ≥10 patients with Primary PCI and ≥10 patients with Fibrinolytic Therapy

Pinto DS … Gibson CM, Circulation 2006Pinto DS … Gibson CM, Circulation 2006


Mortality Increases with Increasing PCI Related DelayMortality Increases with Increasing PCI Related Delay

PCI Related Delay (DB-DN) (min)

In H

ospi

tal M

orta

lity

(%) For every 30 minute delay,For every 30 minute delay,

mortality increases 10%mortality increases 10%



0.8

1.25

1.5

Estimated Treatment Effect and 95% Confidence IntervalsConditional on PCI Related Delay (DB-DN)(min)

After Adjusting for CovariatesEs

timate

d Trea

tmen

t Effe

ct

0.5

1.0

2.0

PCI Related Delay (DB-DN) (min)60 75 90 105 120 135 150 165 180114

Odd

s of

Dea

th w

ith

Fibr

inol

ysis


PCI B

ette

rFi

brin

olys

is B

ette

r

The Advantage of PCI Compared with Fibrinolyisis Decreases as the PCI Related The Advantage of PCI Compared with Fibrinolyisis Decreases as the PCI Related Delay IncreasesDelay Increases


0.8

1.25

1.5

Estimated Treatment Effect and 95% Confidence IntervalsConditional on PCI Related Delay (DB-DN)(min)

After Adjusting for CovariatesEs

timate

d Trea

tmen

t Effe

ct

0.5

1.0

2.0

PCI Related Delay (DB-DN) (min)60 75 90 105 120 135 150 165 180114

Odd

s of

Dea

th w

ith

Fibr

inol

ysis


PCI B

ette

rFi

brin

olys

is B

ette

r

Pinto DS … Gibson CM. Circulation. 2006Pinto DS … Gibson CM. Circulation. 2006 *Betriu A. Am J Cardiol. 2005; 95:100-101.*Betriu A. Am J Cardiol. 2005; 95:100-101.

Randomized Studies*


Prehospital Delay & Timing of Reperfusion Strategy Prehospital Delay & Timing of Reperfusion Strategy EquivalenceEquivalence

0-120

121+NonAnt MI65+ YRS Ant MI 65+

YRS NonAnt MI<65 YRS Ant MI <65

YRS

179

168 148

107 103

58 43

400

60

120

180

Prehospital Delay (min)

19,517

5,296

9,812

41,774

16,119

20,424

10,614

PCI R

elat

ed D

elay

(DB

-DN

) Whe

re

PCI a

nd F

ibrin

olyt

ic M

orta

lity

Are

Equ

al (M

in)

3,739



Pre-hospital Delay (Min)

Age <65 years & Anterior Infarction

(n=patients, n=hospitals)

Age <65 years & Non-anterior Infarction


Age ≥65 years & Anterior Infarction


Age ≥65 years & Non-anterior Infarction


0-120 39 minMet goal without transfer: 2.6%

Met goal with transfer: 3.6%

(n=19,517 pts; n=269 hospitals)

56 minMet goal without transfer:

20.2%Met goal with transfer:

17.3%(n=41,774 pts; n=385 hospitals)

109 minMet goal without transfer: 94.1%



154 minMet goal without transfer: 99.8%Met goal with transfer: 100.0%(n=20,424 pts; n=271 hospitals)

121+ 50 minMet goal without transfer: 7.9%Met goal with transfer: 11.9% (n=5,296 pts; n=117 hospitals)


Met goal with transfer: 82.2%(n=16,119 pts; n=244 hospitals)





Met goal with transfer: 100.0%(n=10,614 pts; n=191 hospitals)

Time At Which PCI Looses Superiority In Survival Over Time At Which PCI Looses Superiority In Survival Over Fibrinolysis Varies Depending Upon Patient Risk Fibrinolysis Varies Depending Upon Patient Risk

Pinto DS … Gibson CM, Circulation Pinto DS … Gibson CM, Circulation 20062006


Time of OnsetTime of Onset

ED Time Point 1: ED Time Point 1: DOORDOOR

ED Time Point 2: ED Time Point 2: DATADATA

ED Time Point 3: ED Time Point 3: DECISIONDECISION

ED Time Point 4: ED Time Point 4: DRUGDRUG

Time Interval IIIDecision to drug

Time Interval IIECG to decision to treat

Time Interval IDoor to ECG

NHAAP Recommendations. U.S. Department of Health NIH Publication: 1997:97-3787.

The Four DsThe Four Ds


75

105

135

165

195

225

255

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Min

utes

(Med

ian)

Min

utes

(Med

ian)

NRMI 1NRMI 1 NRMI 2NRMI 2 NRMI 3NRMI 3 NRMI 4NRMI 4

NRMI Non-Transfer-In PatientsNRMI Non-Transfer-In Patients

NRMI Transfer-In PatientsNRMI Transfer-In Patients

Door to Balloon Times By Transfer Status, Primary PTCA PatientsDoor to Balloon Times By Transfer Status, Primary PTCA Patients

228

111

171

100110: DANAMI Transfer

90: DANAMI On Site


05

1015202530354045

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Perc

ent o

f Pat

ient

sPe

rcen

t of P

atie

nts


Non-Transfer-In PatientsNon-Transfer-In Patients

Transfer-In PatientsTransfer-In Patients

Door to Balloon Times Door to Balloon Times <<90 Minutes By Transfer Status, Primary 90 Minutes By Transfer Status, Primary PTCA PatientsPTCA Patients

3.8

33.6

39.1

5.3


20

25

30

35

40

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Perc

ent o

f Pat

ient

sPe

rcen

t of P

atie

nts


All Patients

Door to Balloon Times Door to Balloon Times <<90 Minutes90 Minutes

34.5 %

29.4 %


National Trends in AMI Management:National Trends in AMI Management:Door to Drug Time with ThrombolysisDoor to Drug Time with Thrombolysis

0102030405060708090

100

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

Med

ian

time,

min

utes

NRMI 1 NRMI 2 NRMI 3 (Activase only) (All lytics) (All lytics)

60

91

34 39

NRMI 1: Includes patients where initial ECG was the method of MI diagnosis NRMI 2 and 3: Includes patients with ST on 1st 12-lead ECG results, where 1st 12-lead ECG date/time = 1st 12-lead ECG with ST and/or BBB date/time

Non-transfer-in patients

75th percentile, 52

25th percentile, 22


National Trends in AMI Management:National Trends in AMI Management:Door to Balloon Time in PPTCADoor to Balloon Time in PPTCA

80

90

100

110

120

1994 1995 1996 1997 1998 1999

Med

ian

time,

min

utes NRMI 2 NRMI 3

116

108

Includes patients with ST on 1st 12-lead ECG results, where 1st 12-lead ECG date/time = 1st 12-lead ECG with ST and/or BBB date/time (non-transfer-in patients)

Time Dependence of Reperfusion Therapy: “Wavefront” Phenomenon of Necrosis in Dogs

Documents

Transcript of Time Dependence of Reperfusion Therapy: “Wavefront” Phenomenon of Necrosis in Dogs