Bivalirudin Reduces Cardiac Mortality in Patients with and Without Major Bleeding

Gregg W. Stone, Tim Clayton, Roxana Mehran, Efthymios N. Deliargyris, Jayne Prats, Stuart J. Pocock TCT 2012; JACC 2012;60(17SupplB):B16

The HORIZONS-AMI Trial

Bivalirudin Reduces Cardiac Mortality in Patients with and Without Major Bleeding

Background

● In the HORIZONS-AMI trial, treatment with bivalirudin compared to heparin + a GPIIb/IIIa inhibitor in patients with STEMI undergoing primary PCI resulted in markedly reduced rates of cardiac mortality, which is usually attributed to decreased bleeding

● Whether the reduction in mortality with bivalirudin can be fully ascribed to reduced bleeding is unknown

Stone GW et al. TCT 2012 and JACC 2012;60(17SupplB):B16

Cardiac Mortality30 days to 3 years*

'

Bivalirudin (n=1800)Heparin + GPIIb/IIIa (n=1802)

Stone GW et al NEJM 2008;358:2218-30; Mehran R et al Lancet. 2009;374:1149-59; Stone GW et al. Lancet 2011;377:2193-204

Car

diac

Mor

talit

y (%

)

3-yr† HR [95%CI]=0.56 [0.40, 0.80]

P=0.001

2.9%

5.1%1-yr† HR [95%CI]=0.57 [0.38, 0.84]

P=0.005

0 12 15 18 21 24 27 30 33 36

Months3 6 9

0

1

6

5

4

3

2

3.8%

2.1%

30-d† HR [95% CI] 0.62; [0.40,0.96]

P = 0.03

1.8%

2.9%

*All cause mortality at 3 years was also consistently lower with bivalirudin (5·9% vs 7·7%), HR 0·75 [0·58–0·97]; p=0·03 †These timepoints were prespecified analyses

Risk factor Hazard ratio (95% CI) P-value

Age (per 5 years) 1.31 (1.21 to 1.43) <0.001

WBC (per 109 cells/L) 1.12 (1.07 to 1.18) <0.001

Creatinine (per 0.1 mg/dL) 1.11 (1.06 to 1.16) <0.001

Killip class 2-4 2.41 (1.62 to 3.60) <0.001

LAD PCI 1.68 (1.16 to 2.45) 0.007

Diabetes, medically treated 1.54 (1.06 to 2.23) 0.02

Bivalirudin (vs UFH+GPIIb/IIIa) 0.57 (0.40 to 0.81) 0.001

3-Year Cardiac Mortality

Other variables in model: current smoker, female gender, prior MI, # vessels treated, hemoglobin


● In all patients, multivariable model

12

0

4

6

8

10

2

012 15 18 21 24 27 30 33 36

3-yr HR [95%CI]=0.64 [0.51, 0.80]

P<0.001

6.9%

10.5%M

ajor

Ble

edin

g, n

on-C

AB

G (%

)

Months3 6 9

Δ=64major bleeds

3-Year Major Bleeding*

* Intracranial intraocular, retroperitoneal, access site bleed requiring intervention/surgery, hematoma ≥5 cm, Hgb ↓ ≥3g/dL with or ≥4g/dL w/o overt source; reoperation for bleeding; or blood product transfusion

Stone GW et al. Lancet 2011;377:2193-204

Bivalirudin (n=1800)Heparin + GPIIb/IIIa (n=1802)

HR [95%CI] = 5.81 [3.92, 8.62]

P<0.001

3.3%

11.6%

Years

Car

diac

mor

talit

y (%

)

12% No major bleed (n=3296)Major bleed (n=306)

10%

8%

6%

4%

2%

0%

0 1 2 3

Impact of Major Bleeding


% major bleed in patients with cardiac death

14.0%(7/50)

30.7%(27/88)

P=0.03

Heparin + GPI Bivalirudin0

5

10

15

20

14.6

5.83.8

2.6

Major bleeding No major bleeding

3-Y

ear C

ardi

ac M

orta

lity

(%)

HR [95%CI] = 4.62 [2.04, 10.45]

P=0.002

HR [95%CI] = 5.67 [3.59, 8.96]

P<0.0001

7/121 43/167927/185 61/1617

Pint = 0.34

3-year Cardiac Mortality


● In patients with vs without major bleeding

Major bleeding No major bleeding0

5

10

15 14.6

3.8

5.8

2.6

Heparin + GPIIb/IIIa (n=1802) Bivalirudin (n=1800)

3-Y

ear C

ardi

ac M

orta

lity

(%)

27/185 7/121 61/1617 43/1679

HR [95%CI] = 2.56 [1.12, 5.88]

P=0.02

HR [95%CI] = 1.47 [1.00, 2.17]

P=0.048

∆ = 20 deaths ∆ = 18 deaths# fewer cardiac

deaths with bivalirudin

Pint = 0.34



● In patients with vs without major bleeding, according to treatment

HR [95%CI] =0.39 (0.17 - 0.89)

P=0.025

0%

2%

4%

6%

8%

10%

12%

14%

16%

Car

diac

mor

talit

y* (%

)

121 104 94 59Bivalirudin185 151 138 86UFH + GPIIb/IIIa

0 1 2 3

Years

Heparin + GPIIb/IIIa (n=185)Bivalirudin (n=121)

5.8%

14.6%

3-year Mortality

*From the time of a major bleed


● In patients with major bleeding, according to treatment

Risk factor HR (95% CI) P-value

Age (per 5 years) 1.33 (1.13 to 1.56) 0.001

WBC (per 109 cells/L) 1.23 (1.12 to 1.36) <0.001

Bivalirudin (vs UFH+GPIIb/IIIa) 0.32 (0.14 to 0.78) 0.006


Other variables in model: diabetes, Killip class, LAD treated, hemoglobin, creatinine


● In patients with major bleeding, multivariable model

Baseline Nadir Decrease0

5

10

15 14.3

10.5

3.9

13.8

10.1

3.7


Hgb

(g/d

L)

± 1.9 ± 2.3

P=0.03

± 1.9 ± 2.0

± 1.7 ± 2.0

P=0.08

P=0.31

Hemoglobin Levels


● In patients with major bleeding

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 28 290

5

10

15

20

25

30

35

Median # units(among patients transfused)

UFH + GPIIb/IIIa: 2 [2, 5] Bivalirudin: 3 [2, 5]

P=0.10

Number of RBC units transfused

Num

ber o

f pat

ient

sRBC Transfusions


0%

1%

2%

3%

4%

5%

0 1 2 3

HR [95%CI] = 0.67(0.46 to 1.00)

P=0.046Car

diac

mor

talit

y (%

)

Years


2.6%

3.8%

3-year Mortality

*KM curve with censoring at time of major bleed


● In patients without major bleeding*

Risk factor Hazard ratio (95% CI) P-valueLMS PCI 10.57 (3.76 - 29.70 <0.001

LAD PCI 1.72 (1.10 - 2.69) 0.02

Age (per 5 years) 1.29 (1.16 - 1.43) <0.001

Killip class 2-4 2.86 (1.82 - 4.51) <0.001

S. creatinine (per 0.1 mg/dl) 1.14 (1.08 - 1.20) <0.001

WBC (per 109 cells/L) 1.08 (1.02 - 1.14) 0.009

Diabetes, insulin treated 1.92 (1.01 - 3.65) 0.047

Hemoglobin (per g/dl) 0.86 (0.76 - 0.98) 0.03

Bivalirudin (vs UFH+GPIIb/IIIa) 0.65 (0.44 - 0.97) 0.035

Other variables in model: current smoker, female gender, prior MI, # vessels treated* patients censored at time of bleed



● In patients without major bleeding*, multivariable model

0

5

10

15

20


13.2%

10.1%

P=0.004

3.1%

8.1%

HR [95%CI] = 2.76 [1.85, 4.14]

P<0.001

ThrombocytopeniaYears

Car

diac

mor

talit

y (%

)

Acquired thrombocytopenia (n=404)No thrombocytopenia (n=3053)

0

10%

8%

6%

4%

2%

0%

1 2 3

Acquired thrombocytopenia,*in-hospital

* Nadir platelet count <150,000 in patients w/o baseline thrombocytopenia



5

10

15

20

12.3

2.33.5

2.5

Thrombocytopenia No thrombocytopenia

3-Y

ear C

ardi

ac M

orta

lity

(%)

HR [95%CI] = 1.44 [0.50, 4.12]

P=0.51

HR [95%CI] = 4.36 [2.73, 6.95]

P<0.0001

4/176 39/156028/228 52/1493

9.3%(4/43)

35.0%(28/80) P=0.002% thrombocytopenia in

patients with cardiac death

Pint = 0.006



● In patients with and without in-hospital acquired thrombocytopenia

Thrombocytopenia No thrombocytopenia0

5

10

15

12.3

3.52.3 2.5


3-Y

ear C

ardi

ac M

orta

lity

(%) HR (95%CI) =

5.56 (2.00, 16.67)P=0.0001

HR (95%CI) = 1.41 (0.47 to 1.09)

P=0.12

4/176 39/156028/228 52/1493

Pint = 0.006


Nadir platelet count <150,000 in patients w/o baseline thrombocytopenia


● In patients with and without thrombocytopenia, according to treatment

Both bleeding and thrombocytopenia

Bleeding alone Thrombocytopenia alone

No thrombocy-topenia or bleeding

0

5

10

15

20

25

30

17.3

8.35.6

2.6

3-Y

ear C

ardi

ac M

orta

lity

(%)

Ptrend <0.0001

14/81 17/205 18/323 74/2848

P=0.005P=0.03

% of 123 cardiac deaths

11.4%(n=14)

14.6%(n=18)

13.8%(n=17)

60.2%(n=74)

Interaction between major bleeding and acquired thrombocytopenia*


*Excluding patients with baseline thrombocytopenia


5

10

15

20

25

30

26.0

3.2

9.6

6.38.4

2.12.9 2.3

Bleed + thrombocytopenia Bleed only

Thrombocytopenia only Neither

3-Y

ear C

ardi

ac M

orta

lity

(%)

Ptrend <0.0001

13/50 12/125 15/178 40/1368 1/31 5/80 3/145 34/1480

Ptrend = 0.17

Interaction between major bleeding, thrombocytopenia* and treatment

*Excluding patients with baseline thrombocytopeniaStone GW et al. TCT 2012 and JACC 2012;60(17SupplB):B16


Excludes 145 patients with thrombocytopenia at baseline. Other variables in model: current smoker, female gender, prior MI, # vessels treated, hemoglobin

Risk factor Hazard ratio (95% CI) P-valueAge (per 5 years) 1.34 (1.23 to 1.46) <0.001WBC (per 109 cells/L) 1.15 (1.09 to 1.21) <0.001S. creatinine (per 0.1 mg/dl) 1.10 (1.05 to 1.16) <0.001Killip class 2-4 2.17 (1.41 to 3.35) <0.001LAD PCI 1.68 (1.13 to 2.50) 0.007Diabetes, medically treated 1.50 (1.01 to 2.23) 0.045Major bleeding 2.97 (1.88 to 4.69) <0.001Acquired thrombocytopenia 2.10 (1.36 to 3.24) 0.001Bivalirudin (vs UFH+GPIIb/IIIa) 0.54 (0.38 to 0.79) 0.002


● Multivariable model, including adverse events

Conclusions● In HORIZONS-AMI, treatment with bivalirudin rather than UFH +

GPIIb/IIIa resulted in a marked reduction in cardiac mortality in patients with STEMI undergoing primary PCI– ~Half of the reduction in cardiac deaths with bivalirudin

occurred in patients without major bleeding

● In addition to reducing major bleeding, bivalirudin reduced the occurrence of thrombocytopenia, which contributed to the improved survival in patients with and without major bleeding

● The adverse effects of major bleeding and thrombocytopenia are mitigated in patients treated with bivalirudin rather than UFH + GPIIb/IIIa, and bivalirudin was strongly associated with reduced cardiac mortality even after accounting for bleeding and thrombocytopenia – further studies are required to identify the non-hematolgic benefits of bivalirudin


Bivalirudin Reduces Cardiac Mortality in Patients with and Without Major Bleeding

Documents