Dr. Ted WarkentinProfessor, Depts. of Pathology & Molecular Medicine, and 

Medicine, McMaster UniversityHematologist and Regional Director, Transfusion Medicine, 

Hamilton, Ontario, Canada

ISTH Advanced Course, Cascais, PortugalSat 15 Mar 2014

Heparin‐Induced Thrombocytopenia (HIT)

DisclosuresOrganization DescriptionGlaxoSmithKline research fundingW. L. Gore consulting, research fundingTaylor & Francis royaltiesInstrumentionLaboratory lecture honoraria

Pfizer Canada lecture honorariaLaw firms medical‐legal testimony

• Specific therapeutic recommendations that are not FDA‐labeled indications (treatment of HIT: danaparoid & fondaparinux)

ObjectivesLearning Objectives‐‐ Review:THEME #1   Characteristic timing features of HITTHEME #2   Strong reactivity at buffer controlTHEME #3   Treatment of HIT: Indirect Xa inhibitors vs DTIs

FcRγIIa

Warkentin TE & Greinacher A. Heparin-Induced Thrombocytopenia. 5th edn. CRC Press, Boca Raton, FL, USA 2013.

HIT is Prothrombotic

HIT is Prothrombotic• Both venous and arterial thrombosis• ~50‐70% of HIT patients develop thrombosis

HIT Non‐HIT RR (95%CI) P_____Prox DVT 8/18  (44%) 26/647 (4%)    11 (6, 21) <0.0001Pulm emb 2/18  (11%) 2/647 (0.3%)    36 (5, 241)       0.004VTE  9/18  (50%) 28/647 (4%)       12 (6, 21) <0.0001

Data from orthopedic surgery database (N Engl J Med 1995; Arch Intern Med 2003)

VTE = venous thromboembolism (proximal DVT and/or pulmonary embolism)

HIT “Paradox”

Type ofSurgery

Odds Ratio(& 95% CI)

Risk Reduction(± SD)

GeneralOrthopedicUrologic

ANY TYPE

67% ± 465% ± 775% ± 15

68% ± 3

0.0 0.5 1.0Heparin better Heparin worseOdds Ratio (heparin : control)

Collins et al. N Engl J Med 1988; 318: 1162-73.

Postoperative DVT: UFH vs Placebo

Heparin ↓thrombosis ~68%;

0.1 1 100.32 3.2Log10 scale

UFH reduces clotsBaseline risk

but HIT ↑thrombosis ~12x;THUS, HIT↑thrombosis ~4x (vs never getting heparin)

HIT ↑clots ~4Xvs no UFH given(+ unusually severe clots)

0.1 1 100.32 3.2Log10 scale

UFH reduces clotsBaseline risk

HIT ↑clots ~4Xvs no UFH given(+ unusually severe clots)

5% limb loss in HIT !

HIT: a “Clinical‐Pathologic” Syndrome

• Thrombocytopenia• Thrombosis (>50%, ven > art)• Timing (proximate heparin)

• oTher cause(s) less likely

“Clinical” “Pathologic”

Warkentin, Chong, Greinacher. Thromb Haemost 1998; 79: 1

The 4 T’s (pre‐test scoring system)

4T’s

2 Points 1 Point 0 Point

Thrombocytopenia>50% fall

(nadir >20)

30-50% fall ornadir 10-19; or >50% (surgery)

<30% fall ornadir <10

Timing c/w HITYes (day 5-10); or <d1 (hep 5-30d)

Yes (>d10); or <d1 (31-90d) No (<d4)

Thrombosis Yes Possible No

OTher Dx No Possible Likely

High probability: 6 – 8 pointsModerate probability: 4 – 5 pointsLow probability: 0 – 3 points HIGH NEG PREDICTIVE VALUEWarkentin & Heddle. Curr Hematol Rep 2003;2:148. Lo et al. J Thromb Haemost 2006;4:759

4Ts Scoring System for HIT

HIT: a “Clinical‐Pathologic” Syndrome

• Thrombocytopenia• Thrombosis (>50%, ven > art)• Timing (proximate heparin)

• oTher cause less likely

• Platelet‐activating       anti‐PF4/heparin IgG (“HIT antibodies”)– Positive SRA or HIPA– Strong positive EIA (surrogate for SRA+)

“Clinical” “Pathologic”

Warkentin, Chong, Greinacher. Thromb Haemost 1998; 79: 1

The 4 T’s (pre‐test scoring system)

Two Types of AssaysPlatelet Activation Assays

SRAHIPA

PF4‐dependent ImmunoassaysEIA (ELISA) PaGIA 

instrumentation‐based

PF4/heparin complexes

HIT-IgG antibodies

Radiolabeled 14C-serotonin released released from normal donor platelets

Serotonin-Release Assay (SRA)

Sheridan D, Carter C, Kelton JG. A diagnostic test for HIT. Blood 1986; 67: 27-30.

Washed platelet activation assay

0.1 UFH 0.3 UFH 100 UFH0

Perc

ent S

erot

onin

Rel

ease

Heparin (U/mL)

0

100

90

80

70

60

50

40

30

20

10

typical cut-off

0

Washed with apyrase (preserves reactivity to ADP)Resuspended in buffer (physiological Ca++, Mg++)

↓Inhibitors of HIT Ab-induced plt act’n(IgG, fibronectin)

heparin

PF4PF4/heparincomplex

HIT-IgG

COLOR

conjugated goatAlkaline phosphatase-

antihuman IgG(from serum or plasma)

PF4/heparin-EIA

Add substrate

PolyspecificEIAs detectall 3 classes: IgG, IgA, IgM

Adapted from: Lee & Warkentin. In: Warkentin & Greinacher, eds. Heparin-Induced Thrombocytopenia, 4th edn . New York: Informa, 2007

(IgG-specificEIAs higherspecificity

EIA-IgG/A/M

Higher ODs in the EIAs

of SRA+ (or HIPA+) statusincrease the probability

EIA-IgG/A/M result (OD units): <0.4 0.4-1.0 1.0-1.5 1.5-2.0 >2.0Probability of SRA+ status: <1% ~5% ~25% ~50% ~90%

EIA-IgG

SRAHIPA

HITHIT-T

Iceberg Model

EIA Optical Density (OD) Levels Strongly Predict for Platelet‐activating Antibodies

2050

200

100150

15

10

5

0

15:0

13:1

7:2

1:42:3

0:6

1:1

106:0

8:0

1:23:2

0:5

1:15

2:0

198:0

4:3389%

9:218%

36:13%

304:00%

5:550%

EIA-GTI

SRA <50:>50HIT likelihood%

<0.4 0.4-1.0 1.0-1.4 1.4-2.0 >2.0OD rangeN

umbe

r of p

atie

nts

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0range (0.2 OD increments)

OD = 1.4-2.0 OD >2.0RISK ~50:50 RISK ~90% OR MORE

+ SRA- SRAWarkentin et al. J Thromb Haemost 2008

Clinical Picture of HIT

3 10 20 50 100 200 500 10005

Thrombosis

Nadir platelet count (x 10-9/µL)

VENOUSDVT

lower limbupper limb (CVC)

PEAdrenal vein (hemorrh.) Cerebral venous

(dural sinus)

ANAPHYLACTOID RxPost-iv UFH bolusPost-sc LMWH

Chills/rigors/feverDyspnea/chest painFlushingTransient globalamnesia

ARTERIALLimb > CVA > MI

SKIN NECROSISat sc injectionsites

MICROVASCULARWarfarin necrosis

Venous limb gangreneCentral skin necrosis

DIC

Withoutthrombosis

Clinical Picture of HIT

Numberof cases(arbitraryscale)

Upper‐limb DVT:role of vascular 

injury

Upper-limb DVT Frequency in HIT

HIT Controls

Odds RatioPatientpopulation Thrombosis

Thrombosis Rate in:(95% CI)Value

P

14/145(9.7%)

3/484(0.6%)

Patients withcentral line

Upper-limb DVT

17.1(4.9-60.5)

<0.001

Hong et al. Blood 2003;101:3049-51 }

Odds Ratio(95% CI)

All 14 upper-limb DVTs occurred at site of central venous catheter:Right, n=12Left, n=2

“We conclude that a localizing vascular injury (catheter use)and a systemic hypercoagulability state (HIT) interact to explain upper-limb DVT complicating HIT.”

P = 0.011

0/145(0%)

Patients withoutcentral line

Upper-limb DVT <0.001

Adrenal hemorrhage (hemorrhagic necrosis)

Hematologist 3 causes of adrenal hemorrhage

1. HIT 2-3% of HIT; 50% are unilateral50% are bilateral (adrenal failure)

2. APS can be a feature of CAPS3. Sepsis Waterhouse-Friderichsen syndrome

Adrenal vein thrombosis secondary hemorrhage

Warkentin et al. Ann Intern Med 1997;127:804-812

Warfarin-induced Venous Limb Gangrene(HIT→ thrombin; warfarin → ↓↓Protein C)

XX

palpable pulses

Macrothrombosis (DVT) +microthrombosis (venules)

Profound Disturbance in Procoagulant‐

Anticoagulant Balance

1. HIT: ↑↑ Thrombin2. Warfarin: ↓↓PC

Days after Aortic Valve Replacement

Aortic valvereplacementsurgery

Sternal wound infection

PICC line heparin flush followed byrespiratory arrest and bleeding

HITPos

(accidental heparin overdose)

HITPos

HITPos

HITNeg

Resp.arrest

Nadir = 32 x 109/L

0 2 4 28 30 32 34 36 38 40 60 138

CPB

0

50

100

150

200

250

300

350

S.C. UFH5000 U BID

Danaparoid

Pla

tele

t Cou

nt (

x 10

9 /L)

Warkentin. J Crit Illn 2002;17:215.

67‐year‐old Female with Respiratory ArrestPost‐Heparin Bolus

Acute Systemic (Anaphylactoid) Reactions to iv Bolus Heparin

• Onset within 5 ‐ 30 minutes• Chills, rigors, fever• Tachycardia, hypertension• Tachypnea, dyspnea• Chest pain or tightness• Diaphoresis, flushing• Nausea, vomiting, diarrhea• Sudden death• Transient global amnesia

Warkentin TE. In: Warkentin & Greinacher, eds. Heparin-Induced Thrombocytopenia, 5th ed. Boca Raton, FL: CRC Press, 2013

400

200

0-2 0 2 4 6 8 10 12 14 16 18 20 22

Days after Start of UFH

Plat

elet

Cou

nt (x

109 /L

)

DVT

474

427

TIMELINE OF POST-UFH BOLUS CARDIAC ARREST

0518h platelet count = 4271050h UFH 5000 U i.v. bolus given1100h UFH 1600 U/hr i.v. given x 30min1105h Onset bradycardia, severe ↓BP

ECG changes of acute MI;1126h CPR for cardiac arrest1131h Death [No repeat platelet count performed][No post-mortem examination performed]

?5000-U bolus

Day 8

Patient #8 84M

SRA+

Study drug: UFH (5000 U b.i.d.)

UFH

open-label UFH (5000 U b.i.d.)

Warkentin et al. Chest 2013;144:848-58.

Death in ICU TrialDeath 15 min interval

“fatal presumed anaphylactoid reaction”

Interpreting Platelet Counts Post‐Surgery

4 8 12 16

Days after surgery0

0

100

200

300

400P

late

let c

ount

(x10

9 /L)

6 10 14 182

Heparin s.c.5000 U bid

Interpreting Platelet Counts

Colon resection

81 F

Day 3 platelet count ~95IS THIS HIT?

Previous Hx of DVT 10y ago

0

200Plat

elet

Cou

nt

x 10

/L

Pre 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Postoperative Day

9

400

600

800

1000

Mean

+2 SD

-2 SD

Normal Postoperative Platelet Counts(mean + 2 SD)

Platelet Counts After Surgery

Warkentin et al. N Engl J Med 1995;332:1330-5

Postoperative thrombocytosis

Early postoperativethrombocytopenia

>50%↓ e.g., d8 

500200

1 2 3 4Postoperative Day

10

20

30

40

60

Per

cent

0

32%

55%

13%

<1%

50

5

0%

6

0%

Day of Postoperative Platelet Count Nadir

Orthopedic surgery data

Day of Platelet Count Nadir (Surgery = Day 0)

Potentially abnormalon/after day 5

Greinacher & Warkentin. In Marder et al., eds. Hemostasis & Thrombosis. Basic Principles & Clinical Practice, 6th edn. Philadelphia, LW&W 2013.

1 2 3 4Day of Platelet Count Nadir (Surgery = Day 0)

10

20

30

40

60

Per

cent

0

50

5 6

Day of Postoperative Platelet Count Nadir

Comparison:orthopedic vs cardiac

Potentially abnormalon/after day 5

Greinacher & Warkentin. In Marder et al., eds. Hemostasis & Thrombosis. Basic Principles & Clinical Practice, 6th edn. Philadelphia, LW&W 2013.

4 8 12 16

Days after surgery0

0

100

200

300

400P

late

let c

ount

(x10

9 /L)

6 10 14 182

Heparin s.c.5000 U bid

Interpreting Platelet Counts

Colon resection

81 F

Day 9 platelet count ~20IS THIS HIT?

Platelet count fall on day 5 of heparin(first day of heparin= day 0)

Day 3

nadir

Timing of HIT

Rapid

Typical onset

Previous HeparinExposure to

UnlikelyPossibleDefinite

1 2 3 4 5 6 7 8 910111213Days after Heparin Exposure

20

60

80

0

40

Num

ber o

f Pat

ient

s

onset

Timing of Typical-Onset HIT

Warkentin & Kelton. N Engl J Med 2001;344:1286-1292

All had previous heparin exposurewithin last 100 days

HIT Antibodies are Transient

Enzyme-immunoassay

Serotonin release assay

1.0

0.8

0.6

0.4

0.2

00 25 50 75 100 125

Days to Negative Assay Result

Freq

uenc

y of

Rep

eat

Pos

itive

Tes

t for

HIT

-Abs

P=0.0073

Warkentin & Kelton. N Engl J Med 2001

0

50

100

150

200

250

-2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

Platelet count fall beganon day 6 of heparin treatment

Platelet count nadiron day 11 (60 x 109/L)

UFH

Abrupt fall in platelet count from179 to 49 x 109/L with repeat use of heparin (day 30)

Days after Starting Heparin

5,000 U bid sc + i.v. infusion5,000 U bolus

Typical onset Rapid onsetof HIT of HIT

Warkentin & Kelton. N Engl J Med 2001;344:1286

Plat

elet

Cou

nt (x

109 /L

)

Timing of Onset of HITRapidTypical

Antibodies newly formed

Timing day 5 to 10irrespective of historyof previous heparin

Timing immediate (<24 h)

Antibodies already present

Recent heparin(past 100 days) crucial

What is the explanation for this speedy (5d) “primary” immunization?

bacteria

platelets

PF4

B-cell

polyanions

granulocyte

heparin

PF4B-cell

granulocyte

anti-PF4/polyanion IgG

activated platelet

HIT

One antibody specificity recognizes a large variety of bacteria = innate humoral immune defenseHIT is a misdirected host defense

4 8 12 16

Days after surgery0

0

100

200

300

400P

late

let c

ount

(x10

9 /L)

6 10 14 182

Heparin s.c.5000 U bid

Why do Platelets Fall After Stopping UFH?

Colon resection

81 F

Day 9 platelet count ~20IF THIS IS HIT,

why are plateletsfalling off heparin?

0.1 UFH 0.3 UFH 100 UFH0

Perc

ent S

erot

onin

Rel

ease

Heparin (U/mL)

0

100

90

80

70

60

50

40

30

20

10

typical cut-off

0

Delayed‐Onset HIT

Delayed‐onset HIT:Definition

platelet count begins to fall,1 or continues to fall,2 despite 

stopping all heparin1 Warkentin & Kelton. Ann Intern Med 2001; 135: 5022 Warkentin. Hematol/Oncol Clin N Am 2010; 24: 755.

0 2 4 6 8 10 12 14 16

0

300

200

100

Platelet Cou

nt (x10

9 /L)

UFH (intra‐ or peri‐op)

± UFH or LMWH prophylaxis(“delayed‐onset HIT” if off heparin)

0 2 4 6 8 10 12 160

3

2

1

HIT‐Ab (OD)

Macrovascular to Microvascular Thrombosis

14

and PF4 release (vicious cycle)Progressive platelet activation

Consumptive coagulopathy

INR , PTT  ,  fibrinogen

Protein C pathway activation

Immunizing heparin exposure

Postoperativethrombocytopenia,day 1‐4 (hemo‐dilution, plateletconsumption)

intensifies during 2nd week

Days after Heparin ExposureWarkentin. Hematol/Oncol Clin N Am 2010; 24: 755‐75.

Hemodilution

0 5 10 15 20 25 30 40

80,000

140,000

160,000

180,000

35

40,000

120,000

0

20,000

100,000

Days after Starting Heparin

Delayed-Onset HIT

Warkentin & Bernstein. N Engl J Med 2003: 348: 1067-9.

Pla

tele

t Cou

nt (p

er m

m3 )

7,000/mm3

Heparin 5,000 U (preoperative)once by subcutaneous injection

Left-lower limb proximal DVTand pulmonary embolism

Platelet transfusionpre-inferior vena cavafilter insertion

Gastric bypass surgery

Plateletcounts

Progressivestroke

Pla

sma

Fibr

inog

en (m

g/dL

)

100

300

600

700

0

200

400

500

Plasmafibrinogenlevels

133 mg/dL

HeparinRechallenge(Previous HIT)

• N=20 patients with previous HIT– 0/3 medical pts formed Abs (despite full course of hep!)– 11/17 (65%) surgical pts formed anti‐PF4/H Abs

• 8/11 (73%) anti‐PF4/H Ab+ pts became +SRA– high SRA+ frequency (? memory for plt‐activating Abs)–1/8 pts  recurrent HIT (despite no postop heparin!) HOW IS THIS POSSIBLE?

• Thus, reasonable to consider heparin re‐exposure, especially for cardiac or vascular surgery        (caveat: delayed‐onset HIT remains possible)Warkentin & Sheppard. Blood 2014. [Epub ahead of print]

Heparin Rechallenge

0

100

200

5025

75

Pla

tele

t cou

nt (x

10-9

/L)

125150175

250225

250

0

100

200

5025

75

Days after surgery

Pla

tele

t cou

nt (x

10-9

/L)

125150175

0 7

225

14 21 28 35 42 49 56 86

IV IgG

Nadir = 20 (day 10)

Nadir = 26 (day 10)

6371

76

0 7 14 21 28 35 42 49 56 63

IV IgG

A

B

Fondaparinux, therapeutic-dose (7.5 mg per day)(2.5 mg/day)

Onset

Onset

Danaparoid, therapeutic-dose with transition to warfarin therapy

Cardiac surgery (heparin) exposure

1st Episode of HIT (1998)

2nd Episode of HIT (2009)

Cardiac surgery (heparin re-exposure)

Fondaparinux

of HIT,day 7DVT and PE

of HIT,day 7DVT by US

Two Episodes of HIT

Warkentin & Sheppard. Blood 2014. [Epub ahead of print]

OD cutoff <0.45

0

100

80

60

40

20

0

3.0

2.0

1.5

1.0

0.5

2.5

0 5 10 15 20 25 30

0

Ser

oton

in re

leas

e, p

erce

nt

0

100

80

60

40

20

EIA

-IgG

, EIA

-IgA

, EIA

-IgM

(OD

uni

ts)

00.3

100 00.1

1.20.41000.8

UFH (IU/mL) Fonda (μg/mL)S

erot

onin

rele

ase,

per

cent

Days after surgery

neat 1/8 1/16 1/32 1/64 1/128 EIA-IgG EIA-IgA EIA-IgM

DC2nd Episode of HIT 2nd Episode of HIT

(day 10)

Day 6 EIA-IgGseroconversion

Day 6 SRA seroconversion

Day 7 EIA-IgMseroconversion

SRA+ on Day 6, Delayed-onset HIT Abs, No Fx X-Reactivity

Warkentin & Sheppard. Blood 2014. [Epub ahead of print]

Previous HIT Episode Heparin Rechallenge

3

8

15 * *

G A M G A MWeeks to

rechallenge

132

180

515

16 422

11 * ** 47

10 37

Patientnumber

0.45 - 0.99 1.00 – 1.99 ≥2.00Antibody OD:

20 20

Wanaka et al. J Thromb Haemost 2010

Typical‐onset HITRapid‐onset HITDelayed‐onset HITPersisting HIT

Spontaneous HIT

160

140

120

100

80

60

40

20

0

Days after cardiac surgery

Platelet cou

nt (x10

‐9/L)

UFH (70,000 IU)

0 2 10 12 164 6 8 14 18 20 28 3022 24 26FondaparinuxEnoxaparin

4436Rivaroxaban

8879 126112

Onset of HIT, d6100

Platelet count nadir = 13 d11

Cardiac surgery

Kopolovic & Warkentin. CMAJ 2014 in press.

PERSISTING HIT, i.e.,Duration of HIT >30 days

160

140

120

100

80

60

40

20

0

Days after cardiac surgery

Platelet cou

nt (x10

‐9/L)

UFH (70,000 IU)

0

100

80

60

40

20

0 0.1 0.3 100UFH (IU/mL)

Serotonin‐release (percent)

Results of SRA(d8 serum)

1/41/81/161/321/641/1281/256

dilutionserum

neat

0 2 10 12 164 6 8 14 18 20 28 3022 24 26FondaparinuxEnoxaparin

d18d8 d11 d14

9376

9281

9262

8953

4436

d44

1095release with (mean 0.1, 0.3) U/mL UFH

sample day

Rivaroxaban

Percent release (1/4 serum dilution) Percent release (neat)

release with 0 U/mL UFH

8879 126112

d88

394

Percent release (neat)

Onset of HIT, d61000 0.1 0.3 100

UFH (IU/mL)0 0.1 0.4 0.8 1.2 10

Results of SRA(d11 serum)

Fondaparinux (μg/mL)

1/41/161/641/256

dilutionserum

fondaparinuxcross‐reactivity

Negative for

Platelet count nadir = 13 d11

Cardiac surgery

Kopolovic & Warkentin. CMAJ 2014 in press.

Per cent release at 0 U/ml UFH(“buffer control”) inverselyproportional to platelet counts

Spontaneous HIT(or Autoimmune HIT)

DEFINITIONdisorder mimicking HIT both clinically and serologically 

except for no proximate heparinWarkentin et al. Am J Med Med 2008; 121: 632. Jay & Warkentin. J Thromb Haemost 2008; 6: 1598.Pruthi et al. J Thromb Haemost 2009; 7: 499. Warkentin et al. Blood 2014; in press.

0 1 2 3 4 5 6 7 8 9 13 14 15

0

50

100

150

200

250

Days after Admission for Stroke

Pla

tele

t Cou

nt (x

10-9

/L)

10 11 12

62-y.o. male admitted for acute thrombotic stroke Platelet count = 65 x 109/LNo recent hospitalizations, no previous heparin

Mechanical thrombectomyIntra-arterial t-PA (15 mg)UFH 1000 IUComplicated by multiple rethrombosesrequiring multiple thrombectomies

Platelet count nadir, 27 x 109/L

Fondaparinux7.5 mg SCdaily x 3

Argatroban IVtarget 2-timesbaseline APTT

ASA 325 mg daily x 5 Warfarin

16

Platelet transfusions1U 1U 1U 2U

Serotonin-release assay and enzyme-immunoassays (3 assays): POSITIVE on day 0 and day 14

Spontaneous HIT Syndrome

Warkentin et al. Blood 2014; in press.

0.1 0.3 100 0

% S

erot

onin

Rel

ease

Heparin (U/mL)

0

100

80

60

40

20 typical cut-off

0

Treatment of HIT

• 2 Do’sStop Heparin (LMWH, flushes,…)Give alternative anticoagulant

• 2 Don’tsNo warfarin (vit K if warfarin given)No prophylactic platelet transfusions

• 2 DiagnosticsTest for HIT antibodiesUltrasound for lower‐limb DVT

Danaparoid (not in U.S.)Fondaparinux (off-label)Lepirudin (discontinued)Argatroban (APPROVED)Bivalirudin (off-label)

Adapted from Warkentin TE. Circulation. 2004; Warkentin et al. Chest 2008

Six HIT Treatment Principles

AT3-Dependent Anti-FXa Inhibitors vs DTIsAnti-FXa Inhibitors

(Danaparoid, Fondaparinux)

Half-life

Adapted from: Warkentin TE. Hematology Am Soc Hematol Educ Program 2011

Dosing

MonitoringEffect on INRProtein C pathwayReversibility

Non-HIT indications

Platelet activationDrug clearanceInhibit clot-bound IIaApproved for HIT

Long (fonda ~17h)Prophylactic/therapeuticDirect (anti-FXa levels) Indirect (PTT)

No effect ↑INR (esp. arg)No effect ? Inhibit APC gen’nNo (covalent AT3-Xa) Yes (non-covalent)

Numerous Not established

Inhibits (danap only) No effectRenal HepaticNo YesYes (danaparoid) Yes

DTIs*(Argatroban, Lepirudin)

*Desirudin & bivalirudin

Short (<1h)Therapeutic only

√√√√√√

√

√

√√

Cost Low (fondaparinux) High√are potential options

√

0 2 4 6 8 10 12 14 16

0

300

200

100

Platelet Cou

nt (x10

9 /L)

UFH (intra‐ or peri‐op)

± UFH or LMWH prophylaxis(“delayed‐onset HIT” if off heparin)

0 2 4 6 8 10 12 160

3

2

1

HIT‐Ab (OD)

Macrovascular to Microvascular Thrombosis

14

and PF4 release (vicious cycle)Progressive platelet activation

Consumptive coagulopathy

INR , PTT  ,  fibrinogen

Protein C pathway activation

Immunizing heparin exposure

Postoperativethrombocytopenia,day 1‐4 (hemo‐dilution, plateletconsumption)

intensifies during 2nd week

Days after Heparin ExposureWarkentin. Hematol/Oncol Clin N Am 2010; 24: 755‐75.

14 15 16 17 18

Pla

tele

t cou

nt

Days after immunizing intraoperative heparin exposure

APTT targetrange

0

20

40

60

80

AP

TT (s

ec)

(x10

9 /L)

100806040200

APTTnormalrange

0.500.25

Arg

atro

ban

(µg/

kg/m

in)

Progressive ischemic limb necrosis necessitating amputations

0

Warkentin. Hematol/Oncol Clin N Am2010; 24: 755-75.

PTT Confounding Argatroban for HIT

Pre-argatroban ↑PTT

Argatroban

HIT‐Associated DIC and “PTT Confounding” of Direct Thrombin 

Inhibitor (DTI) Therapy of HIT

Simple Rule:If “baseline” (pre‐

treatment) PTT is ↑, PTT‐based nomogram is unlikely to be successful (“PTT confounding”)

0 4 8 12 16 20 24 26−2

0

100

200

300

Plat

elet

Cou

nt (x

109

L-1 )

Days after Starting Immunizing Heparin

400

−4 2 6 10 14 18 22

0

33

67

100

133

0

1.0

2.0

3.0

4.0

PTT INR

0

2.01.0

Surgery Onset of HIT

81F

166

126

50

80

18 = Platelet count nadir

Bilateral

Admission to ICU

UFH 5000 U

PTT = 35 s

1.0Dalteparin 2500 U once sc,

0.1

Neurologic injury

Hip fracture

1.7

(adrenal crisis)Profound hypotension

DVTWarfarin

bid scArg dosing, mcg/kg/min

then 5000 sc OD

(ULN)

Linkins & Warkentin. 2011;37:653.

PTT Confounding of DTI Therapy

Warkentin:“[For] those patients with severe HIT who evince concomitant DIC, their hypercoagulability state can be ‘untreatable’ with the approved DTIs, at least when employing 

standard PTT monitoring regimens.”Warkentin TE. Exp Opin Drug Safety, 2014 Jan; 13: 25-43.

Avoiding Treatment Failure 

Due to PTT Confounding

5 10 15 25

Days after starting heparin

0

100

300

200

400

heart surgery

Danaparoid sodium…adjusted by anti-Xa levels

350 3020

DICPTT27→ 42Fbgn2.8 → 1.0PSO4neg- >4+

Warfarin given

1.0

0.8

0.6

0.4

0.2

therapeuticrange

Plat

elet

cou

nt x

109

/L (

)A

nti-X

a u/

mL

( )

Plt = 17 (falling)Ischemic feet

Danaparoid held1. Low platelets

2. Procedure

Warkentin Hematol/Oncol Clin N Am 2010

Fondaparinux for HITStudies with >5 Patients and +EIA

N (% withHIT-thrombosis)

MajorBleeding

NewThrombosis

Kuo & Kovacs 2005

Lobo et al. 2007

Grouzi et al. 2009

Pooled data N=60 (67%) 1/60 (1.7%)0/60 (0%)

N=24 (58%) 0/24 (0%)0/24 (0%)

N=7 (86%) 0/7 (0%)0/7 (0%)

N=5 (100%) 0/5 (0%)0/5 (0%)

Warkentin. Hematol/Oncol Clin N Am 2010; 24: 755-75; plus Warkentin et al. & Goldfarb & Blostein. J Thromb Haemost 2011 (Dec)

Goldfarb & Blostein 2011* N=8 (75%) 0/8 (0%)0/8 (0%)

Warkentin et al. 2011** N=16 (56%) 1/16 (6%)0/16 (0%)

* All 8 patients had positive SRA or strong positive EIA (>2.00 OD units)** All 16 patients had positive SRA (mean EIA = 2.53 OD units)

Prevention of HIT

Study

Leyvraz 1991 **

Warkentin 1995 *

Ganzer 1999 *

Pouplard 1999 **

Mahlfeld 2002 *

Total (95% CI)

Risk of HIT: Odds Ratio (95% CI)

Favors LMWH Favors UFH0.001 0.01 0.1 1 10 100 1000

Warkentin. Blood 2005;106:2600 [Commentary on Martel et al. Blood 2005:106:2710].

Meta-Analysis of UFH vs LMWH

* Enoxaparin** Dalteparin

Common odds ratio = 0.10(95% CI, 0.03-0.30)

Preventing HIT in the ICU with 

LMWH(Dalteparin)

PROTECT Trial: main findings(“as-treated”a analysis)

Dalteparin(N = 1827) P

UFH(N = 1832)

Proximal DVT

PE (any)b

Death (in-hosp.) 396 (21.7%) 0.15446 (24.3%)

22 (1.2%) 0.0142 (2.3%)

94 (5.1%) 0.54108 (5.9%)

Warkentin TE. Crit Care Clin 2011 Oct; 27 (4): 805-823, summarizes N Engl J Med 2011; 364: 1305-1314.

Bleeding (major) 100 (5.5%) 0.88105 (5.7%)

HIT 5 (0.3%) 0.1712 (0.7%)

a Excludes patients where consent withdrawn, incorrectly randomized, or study drug not given. b Includes all PE’s classified as: “definite”, “probable” or “possible”c Excludes patients with VTE on study entry; includes patients who received study drug ≥2 d;

and who had ≥ technically-adequate noninvasive imaging for DVT

Outcome

HIT (per-protocolc) 3/1566 (0.2%) 0.04612/1561 (0.8%)

Hazard ratio(95% CI)

0.90 (0.78, 1.04)

0.48 (0.27, 0.84)

0.91 (0.68, 1.23)

0.98 (0.73, 1.31)

0.47 (0.16, 1.37)

0.27 (0.08, 0.98)

6/157(3.8%)

0/171(0.0%)

3/104(2.9%)

1/201(0.5%)

2/176(1.1%)

7/1502(0.5%)

11/437 (2.5%)

80% reduction95% CI2.19, 17.34p<0.0001

8/1874 (0.4%)

1996 1997 1998 1999 2000 2001 2002 2003

May’96-Jun’97 Apr’98-Dec’99 Jan’00 ---------- Dec’03

Heparin Use and HIT Post-Cardiac Surgery

UFH

LMWH

8/1703(0.5%)

ObjectivesLearning Objectives‐‐ Review:THEME #1   Characteristic timing features of HITTHEME #2   Strong reactivity at buffer controlTHEME #3   Treatment of HIT: Indirect Xa inhibitors vs DTIs