J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 5 , N O . 2 0 , 2 0 1 5

ª 2 0 1 5 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N DA T I O N I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 5 . 0 3 . 5 6 7

THE PRESENT AND FUTURE

STATE-OF-THE-ART REVIEW

Bleeding Avoidance Strategies DuringPercutaneous Coronary Interventions

Mandeep Singh, MD, MPH

ABSTRACT

Fro

rep

Lis

Yo

Ma

Bleeding avoidance strategies for percutaneous coronary interventions continue to evolve with the availability of newer

antiplatelet and anticoagulation therapies. Advances in interventional practices have altered the balance between

ischemic and bleeding complications. With the availability of rapidly-acting platelet adenosine diphosphate–receptor

antagonists, the need for routine glycoprotein IIb/IIIa inhibitors has diminished. Recent meta-analyses and trials have

advanced our knowledge of vascular access and different anticoagulation regimens. Vascular closure devices have

long been used for early ambulation; however, more recent results demonstrating lower bleeding complications

from observational registries are encouraging. This review synthesizes this information, taking into account changes

in the landscape of interventional practice with respect to current bleeding avoidance strategies. (J Am Coll Cardiol

2015;65:2225–38) © 2015 by the American College of Cardiology Foundation.

B leeding and vascular complications in pa-tients undergoing percutaneous coronaryinterventions (PCIs) are associated with

significant costs, prolonged hospital stays, and in-creased short- and long-term morbidity and mortality(1–5). The risk of bleeding is modifiable, and im-proving bleeding and vascular complication rates pro-vides an opportunity to improve the health care andsafety of PCI. In that regard, the Centers for Medicareand Medicaid Services (6) have identified bleedingand hematoma following cardiovascular proceduresas quality indicators (7).

Marso et al. (8) used the term “bleeding avoid-ance strategies” to highlight the importance ofbivalirudin and vascular closure devices (VCDs) inreducing bleeding, using data on more than 1.5million patients undergoing PCI at hospitalsparticipating in the National Cardiovascular DataRegistry (NCDR) (8). In high-risk patients, the use ofboth bivalirudin and VCD was associated withsignificantly lower bleeding rates. Since the publi-cation of this study, pharmacotherapy and

m the Division of Cardiovascular Diseases, Department of Internal Medici

orted that he has no relationships relevant to the contents of this paper

ten to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Vale

u can also listen to this issue’s audio summary by JACC Editor-in-Chief D

nuscript received November 19, 2014; revised manuscript received March

technological advances have shed new light onfactors that can further mitigate bleeding risk inpatients undergoing PCI.

With this backdrop, this review will report recentadvances associated with meaningful reduction inbleeding complications following PCI. It will also re-view the current data on the status of bivalirudin andVCD. Last, this review will provide the reader with apractical strategy to help individualize a patient’sbleeding risk and deploy interventions to reducebleeding in high-risk patients.

DEFINITION OF BLEEDING

Bleeding complications have been identified as acrucial endpoint to test the safety and efficacy of newantithrombotic drugs, cardiac devices, or PCI.Reduction in bleeding events is associated withimproved survival, and prevention of major bleedingmay represent an important step in improvingoutcomes by balancing the safety and efficacy ofpharmacotherapy and devices used during PCI (9).

ne, Mayo Clinic, Rochester, Minnesota. Dr. Singh has

to disclose.

ntin Fuster.

r. Valentin Fuster.

31, 2015, accepted March 31, 2015.

FIGURE 1 Outline of Bleeding

Type 0: No bleeding

Type 1: Bleeding is not actio

Type 2: Any overt, actionabof hemorrhage

Type 3:Type 3a: Overt bleeding pluhemoglobin drop 3 to <5 g/any transfusion with overt b

Type 3b: Overt bleeding pluhemoglobin drop ≥5 g/dl.Incardiac tamponade and blerequiring surgical interventvasoactive agents

Type 3c: Intracranial or intrbleed compromising vision

Type 4: CABG-related bleed Perioperative intrac Reoperation after cl the purpose of cont Transfusion of ≥5 U red blood cells with Chest tube output ≥

Type 5: Fatal bleeding 5a: Probable 5b: Definite

BARC

BARC ¼ Bleeding Academic Res

for Occluded Coronary Arteries;

ABBR EV I A T I ON S

AND ACRONYMS

ACS = acute coronary

syndrome(s)

ACT = activated clotting time

BARC = Bleeding Academic

Research Consortium

CABG = coronary artery

bypass graft

GP = glycoprotein

NCDR = National

Cardiovascular Data Registry

PCI = percutaneous coronary

intervention

VCD = vascular closure de

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2226

Lack of unanimity in defining bleeding hasled to variation in the incidence of bleedingacross institutions performing PCI (1). Thedefinition of bleeding continues to evolve,and 4 such definitions are highlighted inFigure 1. One definition, coined by theBleeding Academic Research Consortium(BARC), classifies bleeding into 5 clinically-meaningful categories (10). This definitionnot only captures the cause (procedural ornonprocedural) of bleeding, but also reflectson the severity, site, and prognostic impli-cations (6). BARC also gives due consider-ation to coronary artery bypass graft (CABG)–related bleeding, as up to 12% of patients

presenting with acute coronary syndromes (ACS) mayundergo CABG during index hospitalization. The keydeterminants of CABG-related bleeding include:duration (24 h for chest tube and 48 h for intracranialbleed and transfusions); site (intracranial); and needfor reoperation. The second definition proposed bythe NCDR is being used to calculate the bleeding riskscore (11). This definition encompasses a broader

vice

Definitions From Different Studies

nable

le sign

sdl andleeding

scludes

edingion or

aocular

ingranial bleeding within 48 hosure of sternotomy forrolling bleedingwhole blood or packedin 48 h2L within 24 h

# Arterial access sitebleeding: •External •Hematoma >10 cm for femoral >5 cm for brachial >2 cm for radial

# Retroperitoneal,gastrointestinal, orgenitourinary bleeding;intracranial hemorrhage;cardiac tamponade

# Post-procedure hemoglobindecrease of 3 g/dl. Inpatients with a pre-procedurehemoglobin level ≤16 g/dl orpost-procedure non-bypasssurgery-related bloodtransfusion for patients with apre-procedure hemoglobinlevel ≥8gm/dl

NCDR

earch Consortium; CABG ¼ coronary artery bypass graft; GUSTO ¼NCDR ¼ National Cardiovascular Data Registry; TIMI ¼ Thrombol

range of access and nonaccess bleeding and accountsfor pre-procedure hemoglobin. Previously, bleedingwas largely defined by variables included in the TIMI(Thrombolysis In Myocardial Infarction) and GUSTO(Global Utilization of Streptokinase and Tissue Plas-minogen Activator for Occluded Coronary Arteries)trials (12,13). Figure 1 highlights the significant het-erogeneity in definitions of bleeding. Such variationsmake the adjudication of endpoints, intertrial com-parisons, and care process improvements difficult, ifnot impossible.

There are still some unresolved issues. First, cli-nicians still must use the NCDR definition to calcu-late the bleeding risk from PCI, and BARC or otherdefinitions need to be utilized to judge severity.Second, the NCDR definition is for bleeding within72 h of PCI and is mainly geared to capture PCI-related complications, whereas the BARC definitionmay need to be used to calculate the short- andlong-term bleeding risk. Third, at present, accessand nonaccess bleeding are not well-differentiated.Nonaccess-site bleeds are more common in patientswho present with ACS, have an adverse prognostic

Minimal: Overt hemorrhageassociated with a fall inhemoglobin <3 g/dl(hematocrit of <9%)

Minor: any clinically overtsign of hemorrhageassociated with a fall inhemoglobin of 3 to ≤5 g/dl(or hematocrit 9 to ≤5%)

Major: (1) Intracranial or (2)clinically significant overtsigns of hemorrhageassociated with a drop inhemoglobin of >5 g/dl(hematocrit of >15%)

TIMI

Mild Bleeding that does notmeet criteria for eithersevere or moderatebleeding

Moderate Bleeding thatrequires blood transfusionbut does not result inhemodynamic compromise

Severe or life-threateningeither intracranialhemorrhage or bleeding thatcauses hemodynamiccompromise and requiresintervention

GUSTO

Global Utilization of Streptokinase and Tissue Plasminogen Activator

ysis In Myocardial Infarction.

FIGURE 2 Temporal Trends in Major Vascular Complications

10

8

6

4

2

0Maj

or V

ascu

lar C

ompl

icat

ions

(%)

MayoClinic

NorthernNew England

CathPCIRegistry

WakeForest

8.40

3.10 3.36

1.20

3.50

1.001.96

0.78

P<0.001 for all temporal trends

Older era

Recent era

Review of 4 major studies showing decline in vascular compli-

cations over time. The differences in the incidence among studies

likely reflect use of different definitions of vascular complica-

tions. Reprinted with permission from Dauerman et al. (22).

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2227

connotation, and do not share a similar temporaldecline as access-site bleeds, and their prevalence,predictors, and prognostic importance need to becaptured separately (5,14–16).

INCIDENCE AND PREDICTORS OF

BLEEDING FOLLOWING PCI

INCIDENCE OF BLEEDING COMPLICATIONS. Varyingestimates from several registries’ data reflect hetero-geneity in defining significant bleeding following PCI.Many centers include hematoma in their bleedingdefinition, leading to higher bleeding estimates, ascompared with more stringent definitions that rely ona significant hemoglobin drop of $3 g/dl with orwithout other major complications (e.g., intraocularor cerebral bleed). For example, data from the MayoClinic included hematomas in their definition ofbleeding and reported higher rates of major vascularcomplications (3.5% in the contemporary era between2000 and 2005), as compared with 2.4% in theNCDR, which included blood transfusion/prolongedhospital stay and/or hemoglobin drop $3 g/dl andaccess or nonaccess site bleeding (14,17). There is anurgent need to adopt a universal definition forbleeding that will facilitate studying temporal trendsand interinstitutional comparisons and that will unifyefforts to improve them.

PREDICTORS OF BLEEDING. Doyle et al. (17) foundfemale sex, older age, renal impairment, larger sheathsize, higher activated clotting time (ACT), use ofglycoprotein (GP) IIb/IIIa inhibitors, VCD use, andlonger procedure time to be predictors of higherbleeding among patients undergoing PCI through thefemoral route. Two models from the NCDR identifiedheart failure, peripheral vascular disease, and pre-sentation with ACS as additional variables predictingbleeding (11,18). Identifying predictors associatedwith higher bleeding rates should prompt health careproviders to seek strategies and interventions tolower the risk.

TEMPORAL BLEEDING TRENDS. Risk estimates forbleeding differed among studies, as they useddifferent definitions of bleeding, making interstudycomparisons difficult. Regardless of the definitionused, all studies have demonstrated a remarkabledecline in access-site bleeding following PCI (17,19–22)(Figure 2). The likely reasons for the declineinclude, among others: increase in the use of radialaccess; smaller sheath size for femoral access;bivalirudin use; and judicious utilization of GP IIb/IIIa inhibitors. Nonaccess-site bleeds have notshared similar declines and continue to dominatein patients with ACS (14).

BLEEDING RISK ESTIMATION. The available riskmodels are sparsely used to predict bleeding and tostratify patients into different risk categories. Indi-vidualizing a patient’s bleeding risk may help pro-viders to tailor access and antithrombotic therapy,choose the duration and severity of anticoagulation,and obviate the need for GP IIb/IIIa inhibitors (23).Due to significant overlap in variables predictingbleeding and other major adverse cardiovascularendpoints, bleeding risk models may also help pro-viders predict mortality and other ischemic endpoints(24–26).

Two contemporary bleeding risk models are avail-able. The first model is derived from 3 trials of ACSand comprises 7 easily-obtainable variables (serumcreatinine, age, sex, presentation, white blood cellcount, cigarette smoking, and randomized treat-ment). TIMI (Thrombolysis In Myocardial Infarction)major bleeding rates increased by bleeding risk scoregroups, from 0.4% for those in the lowest-risk groupto 5.8% for those in the highest-risk group (Figure 3)(27). The second model is derived from the NCDRdatabase, is contemporary, and can be used even inpatients undergoing elective PCI (Figure 4) (11).A patient with a higher bleeding risk can be easilyidentified with these models (e.g., elderly womenpresenting with ACS), allowing bleeding avoidancestrategies to be preferentially targeted to these pa-tients (28).

PROGNOSTIC IMPLICATIONS OF BLEEDING

Bleeding following PCI is associated with anincreased risk of adverse cardiovascular outcomes(17,29–33). Higher risks for mortality, MI, and stent

FIGURE 3 Bleeding Risk Model

Serum creatinine(mg/dl)

Age (years)

Sex

White blood cell|count (/l)

Presentation

Current cigarettesmoker

Antithromboticmedications

Heparin or Bivalirudin plus a GPI0

Bivalirudin monotherapy0

Total Score*

Add toscore

Normal biomarkers(elective and NSTEMI)

0

NSTEMI – Raisedbiomarkers

+3STEMI

+6

No0

Yes+4

<500

50-59+3

60-69+6

70-79+9

≥80+13

≥20+8

18-<20+6

16-<18+5

14-<16+4

12-<14+2

10-<12+1

<100

<1.00

1.0-<1.2+2

1.2-<1.4+4

1.4-<1.6+6

1.6-<1.8+8

1.8-<2.0+10

≥2.0+12

Male0

Female+5

This simple, easy-to-use model is derived from 3 trials of acute coronary syndrome. For a patient who has a creatinine of 1.3 mg/dl, is 72 years

of age, is female, has a white cell count of 11 � 109/l, has non–ST-segment elevation myocardial infarction (NSTEMI) without raised bio-

markers, and is a nonsmoker, her risk score would be: 4 þ 9 þ 4 þ 5 þ 1 þ 0 þ 0 ¼ 19, signifying a 2.7% chance of a non-CABG–related

TIMI major bleed within 30 days. If the patient is treated with bivalirudin alone rather than a heparin plus a glycoprotein IIb/IIIa inhibitor,

the total score should be reduced by 6 to 13 points, indicating a 1.4% chance of a non-CABG–related TIMI major bleed within 30 days.

Reprinted with permission from Mehran et al. (27). PCI ¼ percutaneous coronary intervention; STEMI ¼ ST-segment elevation myocardial

infarction; other abbreviations as in Figure 1.

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2228

thrombosis are noted in patients with bleeding, andbleeding avoidance strategies are associated withimprovements in survival (8). The increased risk frombleeding is not limited to the index hospital admis-sion (5); long-term hazard is also noted (6). The un-derlying mechanisms may include: prothromboticstate; abrupt discontinuation of antiplatelet andanticoagulant therapies; increasing the risk of stentthrombosis; greater prevalence of comorbidities inpatients who bleed; anemia; and the effect of bloodtransfusions to treat bleeding (10,34). Stored bloodused for transfusion has: low 2,3-diphosphoglycericacid activity, thereby increasing the oxygen affinityof hemoglobin and decreasing tissue oxygen de-livery (35); decreased red blood cell deformability,leading to increases in osmotic fragility, aggreg-ability, and intracellular viscosity (36); disruptednitric oxide transport (37); prothrombotic effects; and

transfusion-related immunomodulation (38). The as-sociation of blood transfusion with poor prognosisshould warrant caution toward routine blood trans-fusions in anemic, but stable patients (39,40), butshould not deter clinicians from providing this ther-apy if severe anemia is associated with signs ofischemia (34).

ADVANCES IN PHARMACOTHERAPY

The temporal decline in bleeding complicationsfollowing PCI has followed an evolutionary changein the type and intensity of anticoagulation andantithrombotic therapy. Very high bleeding compli-cations were observed during the 1990s in trials thatmandated the use of GP IIb/IIIa inhibitors con-comitant to intensive heparin therapy (41–43). Withthe use of less intense anticoagulation and with the

FIGURE 4 NCDR Bleeding Risk Model

NCDR CathPCI Bleeding Risk ScoreRisk of Bleeding Based on Point Totals from the

NCDR CathPCI Registry Bleeding Risk Score

STEMI

Age (yr)

BMI

Previous PCI

Chronic kidneydisease

Shock

Cardiac arrestwithin 24 hr

Female

Hb

PCI status

ScoreVariable

No

No

No

No

No

No

0

0

0

0

0

0

0

5

<60

20

15

10

Hb <13

Elective

30

Dialysis

Yes

Yes

Yes

Yes

Yes

0

20

15

35

10

20

0

15

10

5

13 ≤Hb <15

Urgent

Mild

20-30

60-70

0

25

40

10

15

Hb ≥15

Emergency/salvage

Moderate

31-39

71-79

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

0.90

1.10

1.30

1.50

1.70

2.00

2.30

2.70

3.10

3.60

4.20

4.90

5.60

6.50

7.50

8.60

9.90

11.40

13.10

14.90

17.00

Totalpoints

Risk ofbleeding (%)

110

115

120

125

130

135

140

145

150

155

160

165

170

175

180

185

190

195

200

205

210

105 19.30

21.80

24.60

27.50

30.70

34.10

37.60

41.30

45.10

49.00

52.80

56.60

60.40

64.00

67.50

70.80

73.90

76.80

79.40

81.80

84.00

86.00

Totalpoints

Risk ofbleeding (%)

This contemporary model is derived from the National Cardiovascular Data Registry (NCDR), is integer-based, and is parsimonious. Reprinted

with permission from Rao et al. (11). BMI ¼ body mass index; Hb ¼ hemoglobin; other abbreviations as in Figure 3.

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2229

advent of dual antiplatelet therapy in the late1990s, bleeding complications were less commonlyobserved.

Dosing errors still occur and are prevalent invulnerable populations at high risk for bleeding. In anobservational analysis of 30,136 patients from theCRUSADE (Can Rapid Risk Stratification of UnstableAngina Patients Suppress Adverse Outcomes WithEarly Implementation of the American College ofCardiology/American Heart Association Guidelines)trial, 42% of patients admitted with non–ST-segmentelevation myocardial infarction (STEMI) were ad-ministered excessive dosing of unfractionated or low-molecular-weight heparin and GP IIb/IIIa inhibitors(44), which was associated with higher bleeding riskand poor outcomes. This underscores the need fornational quality improvement initiatives to reducedosing errors, thereby yielding improvement inoutcomes.

HEPARIN AND ACT. Since the introduction of PCI,intravenous unfractionated heparin has been thecornerstone of antithrombotic therapy. Despite itsuniversal acceptance, there is still controversyregarding the optimal dose and ACT that reflectsequipoise between bleeding and ischemic complica-tions. In the ESPRIT (Enhanced Suppression of thePlatelet IIb/IIIa Receptor with Integrilin Therapy)trial, ischemic endpoints in patients undergoingcontemporary PCI with stent placement did not in-crease with lower ACT levels, at least to a level of200 s; however, a modest correlation with bleedingwas observed at higher ACT levels (45). The currentAmerican College of Cardiology/American Heart As-sociation PCI guidelines agree that current dosing ofheparin is on the basis of empiricism and experiencefrom randomized trials, and the utility of measuredACT levels in current practice remains uncertain (46).ACT between 200 and 250 s with GP IIb/IIIa inhibitors,

FIGURE 5 Compar

Death

CV Death

MI

Stroke

Death/MI/stroke

Majorbleed

Event

5.7

5.1

5.2

1.2

9.3

3.7

CURETrial

ClopidogGroup

Comparison of recen

CURE (Clopidogrel in

PLATO (Study of Plat

with Prasugrel–Throm

permission from Sch

CV ¼ cardiovascular;

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2230

and between 250 and 300 s without, remains standardpractice with unfractionated heparin during PCI,and operators are encouraged to keep the ACT onthe lower end of the recommended range to lowerbleeding complications.

DUAL ANTIPLATELET THERAPY

The PLATO (Platelet inhibition and Patient Outcomes)trial demonstrated that ticagrelor (an orally-activeagent that binds reversibly to P2Y12), as comparedwith clopidogrel (irreversible blockade of P2Y12,delayed onset, needs cytochrome P450), was associ-ated with a 16% relative risk reduction with regardto a composite of cardiovascular death, myocardialinfarction, and stroke (47). Compared with the CURE(Clopidogrel in Unstable Angina to Prevent RecurrentEvents) and TRITON-TIMI 38 (Therapeutic Outcomesby Optimizing Platelet Inhibition with Prasugrel–Thrombolysis In Myocardial Infarction 38) trials(48–50), not only did PLATO demonstrate significantmortality (death from any cause) reduction (4.5%with ticagrelor and 5.9% with clopidogrel), butthe bleeding risk did not increase (major bleeding11.6% vs. 11.2% with clopidogrel) (Figure 5). However,non-CABG bleeding was significantly higher in pa-tients treated with ticagrelor. The newer antiplatelettherapies provide patients with early, reversible,and predictable platelet inhibition, with reduction in

ison of Dual Antiplatelet Trials

rel

6.2

5.5

6.7

1.4

11.4

2.7

CURETrial

PlaceboGroup

0.93(0.81-1.07)

0.93(0.79-1.08)

0.77(0.67-0.89)

0.86(0.63-1.18)

0.80(0.72-0.90)

1.38(1.13-1.67)

RR withClopidogrel

(95% Cl)

3.0

2.1

7.3

1.0

9.9

2.5

TRITON-TIMI38

PrasugrelGroup

3.2

2.4

9.5

1.0

12.1

1.7

TRITON-TIMI38

ClopidogrelGroup

0.95(0.78-1.16)

0.89(0.70-1.12)

0.76(0.67-0.85)

1.02(0.71-1.45)

0.81(0.73-0.90)

1.45(1.15-1.83)

RR withPrasugrel(95% Cl)

4.5

4.0

5.8

1.5

9.8

11.6

PLATOTicagrel

orGroup

t trials with regard to bleeding and ischemic endpoints. With a reduction in i

Unstable Angina to Prevent Recurrent Events) trial included patients who h

elet Inhibition and Patient Outcomes) and TRITON–TIMI 38 (Trial to Assess Im

bolysis in Myocardial Infarction 38) included patients who had acute corona

omig (106). CHAMPION ¼ Cangrelor versus Standard Therapy to Achieve Op

MI ¼ myocardial infarction; RR ¼ risk ratio.

ischemic endpoints and, importantly, without in-creased bleeding complications. Similar results wererecently reported with intravenous infusion of can-grelor, an intravenous, fast-acting, reversible, anddirect-acting P2Y12 inhibitor. Its antiplatelet effectsare immediate and can be maintained with contin-uous infusion. The plasma half-life of cangrelor isapproximately 3 to 5 min, and platelet function isrestored within 1 h after cessation of the infusion. Theuse of cangrelor in patients undergoing PCI wasstudied in 2 phase 3 trials, the CHAMPION (Cangrelorversus Standard Therapy to Achieve Optimal Man-agement of Platelet Inhibition) PCI and CHAMPIONPLATFORM studies (51). Cangrelor was not associatedwith a significant reduction in the primary efficacyendpoint in either trial, but was associated withreductions in secondary endpoints, including the rateof stent thrombosis, with no excess severe bleeding.The recent CHAMPION PHOENIX trial, comparingcangrelor with clopidogrel, demonstrated the supe-riority of cangrelor, both for reducing ischemic eventsand for not increasing bleeding complicationsfollowing elective and urgent PCIs (52). Advances indual antiplatelet therapy have led to early, predict-able, and reversible platelet inhibition, obviating theneed for routine GP IIb/IIIa inhibition, and therebyreducing bleeding complications.BIVALIRUDIN AND PCI. Bivalirudin is an intra-venous direct thrombin inhibitor that is used as an

5.9

5.1

6.9

1.3

11.7

11.2

PLATOClopidogrel

Group

0.78(0.69-0.89)

0.79(0.69-0.91)

0.84(0.75-0.95)

1.17(0.91-1.52)

0.84(0.77-0.92)

1.04(0.95-1.13)

RR

0.3

0.3

3.8

-

4.7

0.2

CHAMPION-PHOENIXCangrelor

Group

0.3

0.3

4.7

-

5.9

0.1

CHAMPION-PHOENIX

ClopidogrelGroup

-

1.0(0.52-1.92)

1.0(0.52-1.92)

0.80(0.67-0.97)

0.78(0.66-0.93)

1.5(0.53-4.22)

RR

schemic complications, some increase in bleeding events is seen. The

ad acute coronary syndromes without ST-segment elevation; both

provement in Therapeutic Outcomes by Optimizing Platelet Inhibition

ry syndromes with or without ST-segment elevation. Modified with

timal Management of Platelet Inhibition; CI ¼ confidence interval;

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2231

alternative to heparin in patients undergoing PCI.Trials that compared bivalirudin with heparin foundthat net adverse cardiovascular events, whichincluded ischemic and bleeding endpoints, favoredbivalirudin, with superior or noninferior results(9,53–55). The benefit was driven primarily by lowerbleeding complications. There are several limitationsof the previously-published trials that comparedheparin versus bivalirudin monotherapy for PCI.First, the intensity of antiplatelet and anticoagulationtherapies was disparate and weighted heavily towardthe heparin monotherapy arm (56). This made inter-pretation of the primary endpoint challenging, with 2treatment arms expected to have completely oppositeeffects on the incidence of thrombotic and bleedingcomplications. For example, in patients treated withheparin monotherapy, differential and routine use ofGP IIb/IIIa agents was associated with a higher risk ofbleeding complications. Second, even in more recenttrials, such as the EUROMAX (European AmbulanceAcute Coronary Syndrome Angiography) trial, aninternational, randomized, open-label study thatdemonstrated that bivalirudin, initiated duringtransport for primary PCI in patients with STEMI, wassuperior to heparin, the dose of heparin (100 IU/kg)and concomitant optional use of GP IIb/IIIa inhibitors(69.1%) was high, leading to increased bleedingcomplications in the heparin arm (57). Third, ischemicevents, specifically acute myocardial infarction orstent thrombosis (especially within the first 24 h),were more frequently noted in patients treated withbivalirudin (58). Despite the consistent, but nonsig-nificant increase in ischemic events, the concomitantreduction in bleeding events in the bivalirudin-treated patients favored its use as monotherapy.Fourth, upstream use of dual antiplatelet therapy,especially with newer and faster-acting agents, hasled to reductions in the routine use of GP IIb/IIIaagents, currently at 28% from 41% (14). This has led tonewer trials comparing heparin versus bivalirudinmonotherapies, with optional and lower use of GP IIb/IIIa inhibitors. In HEAT-PPCI (How Effective areAntithrombotic Therapies in Primary PercutaneousCoronary Intervention), among 1,829 patients whounderwent emergent coronary angiography at a sin-gle center, GP IIb/IIIa inhibitor usage was low (13% to15%) and was equal in the heparin and bivalirudin arms.The primary efficacy outcome at 28 days (all-causemortality, myocardial infarction, cerebrovascular ac-cidents, unplanned target lesion revascularization)favored heparin (5.7% compared with 8.7% in thebivalirudin group) (59). The primary safety endpointof major bleeding was similar in the 2 groups. Itshould be noted that the median ACT value in the

HEAT-PPCI trial was lower in the heparin arm (206 s)compared with the bivalirudin arm (246 s) and withother primary PCI trials. In the meta-analyses (58),additional heparin and prolonged duration of bival-irudin did not improve ischemic outcomes; however,in the Swedish Coronary Angiography and Angio-plasty Registry, an additional dose of heparin led toimproved rates of myocardial infarction (60).

BRIGHT (Bivalirudin in Acute Myocardial Infarc-tion vs Heparin and GPI Plus Heparin Trial), studiedbivalirudin in patients undergoing PCI. The trial wasperformed in 82 centers in China and randomized2,194 patients into: bivalirudin and provisional GPIIb/IIIa inhibitors (n ¼ 735); heparin and provisionalGP IIb/IIIa inhibitors (n ¼ 729); or heparin and routineGP IIb/IIIa inhibitors (n ¼ 730) (61). The net adverseclinical endpoint was lower in the bivalirudin group(65 events, 8.8%) than with either heparin and pro-visional GP IIb/IIIa inhibitors (96 events, 13.2%) orheparin and routine GP IIb/IIIa inhibitors (124 events,17.0%). The differences between the groups werealmost entirely driven by the differences in the any-bleeding category. There were no differences inischemic events, including stent thrombosis. In theMATRIX (Minimizing Adverse Haemorrhagic Eventsby TRansradial Access Site and Systemic Imple-mentation of angioX) trial of 7,200 ACS patients un-dergoing PCI randomized to bivalirudin orunfractionated heparin, Valgimigli (62) did not reportsignificant improvement in risk of cardiac events orcardiac events plus major bleeding after takingbivalirudin, as compared with standard care. How-ever, bivalirudin was shown to significantly lower therisk of bleeding complications, especially near thecatheter insertion site, which occurred in 1.7% withbivalirudin versus 2.3% in the standard-care group.Additionally, the bivalirudin group had a significantlylower rate of death, most likely related to the reduc-tion in bleeding complications.

The present data suggest that heparin mono-therapy is sufficient for most patients who undergoPCI with radial access. Patients at high risk forischemic complications or stent thrombosis (bifurca-tion or long lesions, among others) may be treatedwith unfractionated heparin with newer P2Y12 an-tagonists. In contrast, patients with a higher risk forbleeding (older women presenting with ACS andwhere femoral access is used for the intervention)may benefit from bivalirudin.

ACCESS AND BLEEDING COMPLICATIONS

FEMORAL ARTERY ACCESS. Access-site bleedingcomplications predominate, especially in the setting

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2232

of elective PCI. It is, therefore, pertinent to advancemeasures and strategies known to lower bleedingrates from access sites. The femoral artery has tradi-tionally been used for PCI in the United States, andapproximately 80% of PCIs are performed throughthis route (63,64). It also is a preferred route forprocedures that require larger catheters. For ex-ample, operators still prefer the femoral access routefor distal left main, complex bifurcation lesions, androtational atherectomy with large burrs. Trans-catheter aortic valve replacements and other periph-eral vascular procedures are also performed throughfemoral or apical routes due to the need for largecatheters.

Few studies have looked into the advantages ofusing smaller-size sheaths, fluoroscopy, and intra-vascular ultrasound for femoral artery access (65).The data are derived from a single center (17), butobservations from randomized trials and registriesare consistent and support the use of a micropunctureneedle (FEMORIS [Femoral Micropuncture or RoutineIntroducer Study]; 21-gauge vs. 18-gauge needle) (66);fluoroscopy or ultrasound (FAUST [Femoral ArterialAccess With Ultrasound Trial]) (67) to target thecommon femoral artery; smaller sheath size; andearly sheath removal. On the basis of these observa-tions, operators are encouraged to use fluoroscopyand micropuncture needles with ultrasound guid-ance, with the goal of performing safe zone arteri-otomy (access between the inferior epigastric arteryabove and the bifurcation of the common femoralartery below).RADIAL ARTERY ACCESS. Radial artery access hasconsistently demonstrated reduction in bleeding andvascular complications following PCI compared withfemoral artery access in observational registries andrandomized trials (63,68,69). In the randomized trial,RIVAL (Radial Vs Femoral Access for Coronary Inter-vention), the primary outcome of death, myocardialinfarction, stroke, or non-CABG–related bleeding at30 days occurred in 3.7% in the radial access group,which was not different from the 4.0% observed inpatients randomized to femoral access (70). However,if the definition of bleeding was broadened to includelarge hematomas and pseudoaneurysms, significantdifferences favoring radial access were observed be-tween the 2 groups. In the recently-published MA-TRIX trial, 8,404 patients with ACS were randomizedto radial (n ¼ 4,197) or femoral (n ¼ 4,207) access forcoronary angiography and PCI. Major adverse car-diovascular events were noted in 8.8% of patientswith radial access, compared with 10.3% of patientswith femoral access. The net adverse clinical events,including bleeding events, favored patients in the

radial access group: 410 (9.8%) patients with radialaccess had net adverse clinical events compared with486 (11.7%) patients with femoral access. The differ-ence was driven by BARC major bleeding unrelated toCABG surgery (1.6% vs. 2.3%) and all-cause mortality(1.6% vs. 2.2%) (71). The largest observational NCDRalso demonstrated that radial access was associatedwith a significant reduction in bleeding and vascularcomplications (14,62). The most significant reductionin bleeding rates was observed in patients with STEMItreated via radial access. Patients with STEMI repre-sent the highest risk for both ischemic and bleedingcomplications. In some randomized trials and meta-analyses, use of the radial approach for access inprimary angioplasty resulted in reduced mortality(69,70,72). With heterogeneity in trial inclusioncriteria and with the majority of deaths in patientswithout a major bleed, there is a lack of a distinct,mechanistic link between mortality reduction andradial access that was observed in some studies (34).However, these results underscore the need to useradial access in this subgroup, which is the mostprone to bleeding complications following PCI.

In addition to lowering bleeding complications,radial access has cost advantages, promotes earlyambulation and discharge, and is preferred by pa-tients (1,70,73). However, despite these advantages,there is a distinct learning curve for proficiency inusing radial artery access for cardiac catheterizationand PCI. In a pre-specified subgroup analyses fromthe RIVAL trial, the primary outcome was reduced(hazard ratio: 0.49, 95% confidence interval: 0.28 to0.87) in high-volume radial centers, defined as theoperator’s annual median radial PCI volume >146(74). Clinical data from 1,298 facilities reporting to theNCDR show that 49% of facilities in the United Statesperformed #400 PCIs and 26% performed #200 PCIsannually; hence, replicating these results in lower-volume centers may be challenging (75). Procedurallearning, however, continues, with further improve-ment noted in dedicated radial PCI operators whoswitch from transfemoral angiography (76). In arecent study from the NCDR, higher-risk patientswere chosen for transradial PCI with increasedoperator radial PCI volume (77). Despite this, oper-ator proficiency improved, and the threshold forovercoming the learning curve appeared to beapproximately 30 to 50 cases. There is growing pub-lished data on the learning curve for transradial PCIs,and despite study differences, 25 to 80 cases areneeded for novice operators (78). In summary,volume-outcome relationships are evident in learningtransradial PCIs, and available data support higher-volume radial PCI centers to improve PCI outcomes.

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2233

RADIAL ARTERY OCCLUSION. Radial artery occlu-sion is a distinct complication of radial access, and itsincidence ranges from 2% to 10% following trans-radial access (79). To prevent ischemia and reuse thesame site for bypass conduit or cardiac catheteriza-tion, the Society for Cardiovascular Angiography andInterventions radial committee recently recom-mended a standardized anticoagulation protocol, us-ing the lowest-profile sheaths, patent hemostasis(80), or radial compression guided by mean arterypressure techniques (81) to reduce its incidence(82). Similar recommendations were made to reduceradiation exposure and to transition to perform pri-mary PCI.

The RADAR (Predictive Value of Allen’s Test Resultin Elective Patients Undergoing Coronary Catheteri-zation Through Radial Approach) trial (83) did notfind a relationship between functional assessment ofdual-artery circulation to the hand by the Allen testand plethysmography and measures of distalischemia, collaterals between the radial and ulnararteries, and functionality of the hand followingradial access. The Allen test was abnormal in 30% anda D pattern by plethysmography was detected in 40%of 942 screened patients. The investigators suggestedthe feasibility of transradial access across the spec-trum of Allen test results. Further studies are neededto test this proof-of-concept.

VCDs AND BLEEDING COMPLICATIONS. VCDs havebeen demonstrated to reduce time to ambulation andincrease patient comfort (84–86). Recent observa-tional registries and subgroup analyses from ran-domized trials have noted reductions in the incidenceof bleeding complications with VCDs following inva-sive femoral angiography or PCI; however, there is alack of randomized trial data to support their use forthese indications (87–89). Observational registry datafrom NCDR and Blue Cross Blue Shield of Michigandemonstrated lower bleeding risk with VCD use(8,90,91). However, meta-analyses and randomizedtrials comparing bleeding rates following manualcompression or from VCD failed to show lowerbleeding rates with VCD (84,88,92). To accuratelyportray the role of VCDs, a high-quality, adequately-powered randomized trial needs to be performed withintention-to-treat and concealment of allocation ofmanual versus VCD strategy. We also need to accountfor the learning curve and do post-marketing sur-veillance to monitor device complications (93,94).Not only is the use of VCDs associated with compli-cations unique to device deployment (loss of limbcirculation or severe infection) (95), but also devicefailure (1.5% to 20%) is associated with significantly

higher bleeding complications (96). Gurm et al. (90)noted, in a large observational registry, a reductionin hematoma and pseudoaneurysm, but also a sig-nificant increase in retroperitoneal bleeds with theuse of VCDs. Their usefulness to reduce bleeding andvascular complications was attenuated in patientswith low BMI and in those on GP IIb/IIIa agents (90).In a recent study from the Massachusetts Departmentof Health, VCD failed in 3.3% of 23,813 procedures(97). VCD failure was associated with excess risk ofany (7.7% vs. 2.8%) or major (3.3% vs. 0.8%) vascularcomplications, as compared with successful VCDdeployment, underscoring the need for physicians tobe vigilant of predictors of VCD failure (female sex,peripheral vascular disease, and emergency status).These are the same variables that increase the risk ofbleeding, and their cautious use or avoidance wouldbe recommended until additional data are accrued.

ADOPTION OF BEST PRACTICES TO

LOWER BLEEDING COMPLICATIONS

Recent analyses of a nationally-representative U.S.PCI population suggest that 12.1% of all in-hospitalmortality after PCI may be related to bleeding com-plications, and may therefore be modifiable, and thatthe number needed to harm (NNH) calculations sug-gest that the mortality risk associated with bleedingwas greatest in patients at the highest bleeding risk(NNH ¼ 21) or with nonaccess-site bleeding (NNH ¼16) (5). Likewise, the NNH was lowest in patients age75 years or older and in patients with STEMI or lowglomerular filtration rate. These high-risk subgroupsmay have the greatest potential for mortality reduc-tion through bleeding avoidance and should be pref-erentially targeted.

Individualizing and stratifying bleeding risk beforecoronary angiography and intervention is paramount(Central Illustration). Not only can one tailor thechoice of vascular access, but modification of anti-coagulation strategies by pre-treatment with dualantiplatelet therapy and use of bivalirudin to loweraccess- and nonaccess-site bleeds in patients deemedto be higher risk will likely lower their bleedingcomplications. Excessive bleeding is also associatedwith acute kidney injury and ischemic complications;thus, preventing bleeding complications will trans-late into overall improvement in PCI outcomes (98).

Recognition and mitigation of bleeding riskfollowing PCI is a recognized health care priority.Despite consistent observations from contemporaryregistry data documenting reductions in vascular andbleeding complications, there is still a paucity ofsystematic efforts to lower bleeding in these patients.

CENTRAL ILLUSTRATION Bleeding Complications FollowingPCI: Proposed Processes Designed To Lower BleedingComplications Following PCIs

• Participation in NCDR CATH-PCI Registry• Periodic assessment of bleeding avoidance

strategies (BAS)• Monitor ischemic and bleeding

complications• Track complications unique to access

(VCD or radial artery occlusion)

POST

AN

GIO

GRAP

HY/

PCI

• Routine use of bleeding risk tool• Pretreatment with Dual Antiplatelet Therapy

(DAPT) if appropriate

• Use radial access as default strategy• Aim for lower activated clotting time (ACT) levels:

- 200-250 secs with GPIIb/IIIa inhibitors- 250-300 secs with/without GPIIb/IIIa inhibitors

• Use of Bivalirudin for high-risk cases(more data needed)

• Judicious and selective use of GPIIb/IIIa inhibitors• Fluoroscopy, ultrasound, and micropuncture-guided

femoral access• Early sheath removal• Vascular closure device (VCD) for early ambulation

(more data needed to demonstrate efficacy to lower bleeding complications)

DURI

NG

ANGI

OGR

APH

Y/PC

IPR

EPRO

CEDU

RE

Singh, M. J Am Coll Cardiol. 2015; 65(20):2225–38.

ACT ¼ activated clotting time; BAS ¼ bleeding avoidance strategies; DAPT ¼dual antiplatelet therapy; GP ¼ glycoprotein; NCDR ¼ National Cardiovas-

cular Data Registry; PCI ¼ percutaneous coronary intervention; VCD ¼vascular closure device.

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2234

Marso et al. (8) used the available risk model tocategorize patients in a large population undergoingPCI into different tertiles of bleeding risk. In thatstudy, the use of VCD and bivalirudin loweredbleeding rates, especially among patients at thegreatest risk for bleeding. The strategies that loweredbleeding risk were used less commonly in the sub-group with the highest bleeding risk (risk-treatmentparadox). In another study, incorporation of individ-ualized risk estimates in the consent form led toreversal of this paradox through rational increase inthe use of bivalirudin (23). At the Mayo Clinic, pre-treatment with dual antiplatelet agents has reducedthe need for GP IIb/IIIa agents, and a concomitantincrease in the adoption of radial access for PCI has

resulted in reduction of bleeding and vascular com-plications (unpublished data). Reduction of bleedingcomplications with the use of bivalirudin needs to beput in perspective. Recent data from HEAT-PPCI andnow from the BRIGHT and MATRIX trials comparingbivalirudin versus heparin has rekindled the debateon anticoagulation during PCI. It has encouraged theuse of heparin monotherapy with pre-loading of dualantiplatelet therapy. Bivalirudin has demonstratedlowered bleeding and may be advantageous in pa-tients with a high likelihood of bleeding. UnlikeHEAT-PPCI, the BRIGHT and MATRIX trials reportedlower bleeding; hence; there is a need to revisit thevalue proposition of bivalirudin. For example, olderpatients, women, those with renal dysfunction, andthose presenting with ACS are subgroups that willbenefit preferentially from bleeding avoidance stra-tegies, including radial access and bivalirudin. Morestudies need to be performed to lower the acutethrombotic risk with bivalirudin, and readers areencouraged to read the recent meta-analysis on thistopic (58). If ischemic complications in the first 24 hfollowing bivalirudin administration can be lowered,short- and long-term prognoses seem excellent (99).In the Swedish Coronary Angiography and Angio-plasty Registry, an additional bolus dose of unfrac-tionated heparin was associated with a lower rate ofdeath or definite target lesion thrombosis at 30 daysin patients undergoing primary PCI with bivalirudinas the anticoagulant (60). The BRIGHT trial demon-strated the usefulness of prolonged bivalirudin infu-sion in lowering ischemic events among patients withacute myocardial infarction undergoing emergent PCI(61).

The recent observational data on the use of VCD tolower bleeding rates following PCI is encouraging(87,89,90). However, the results of meta-analyses ofrandomized trials are not concordant. At present, theuse of VCD to lower bleeding and vascular complica-tions cannot be recommended, and there is a need foran adequately-powered randomized trial.

The use of the radial artery for access hasincreased, albeit at a slower pace in North America.Most trials, including the recent MATRIX trial, andregistry data support the use of the radial artery tolower access-site bleeding complications. Substantialexpertise is needed to perform PCI through the radialapproach, and the best outcomes have only been re-ported at high-volume centers. The prognostic linkbetween bleeding and survival prompts consider-ation of radial artery access as a default strategy inmost patients, but especially in patients who are athigh bleeding risk (older patients presenting withACS), who will derive the most benefit from this

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2235

change (100). In a large study from the British Car-diovascular Intervention Society database, a 35%reduction in 30-day mortality with radial artery ac-cess for PCI was seen across various risk categories,but was most evident in patients at the highestbaseline bleeding risk (4). Current efforts to traininterventional fellows and consultants will certainlyincrease adoption of this technique, but at the sametime, the role and safety of femoral access should notbe undermined. Readers are encouraged to reviewthe meta-analysis comparing radial and femoral ac-cess for coronary angiography and interventions(70,101). Femoral access is still needed for largersheath sizes, complex left main interventions, rota-tional atherectomy with large burrs, transaortic valvereplacement (TAVR), intra-aortic balloon pump andother assist devices, and abdominal and lower-extremity angiography and interventions. High ratesof bleeding complications are noted in patientsundergoing TAVR. A recent study demonstrated areduction in major bleeding rates with the adoptionof heparin doses according to the ACT value. Majorbleeding was higher in patients treated with aweight-based dosing strategy (33.5% vs. 7.5%,p < 0.001). Multivariate adjustment favored the useof ACT-based dosing of heparin, with a significantlylower incidence of 30-day major bleeding (102).Recently, the use of the radial crossover technique,with a salutary effect on bleeding, has been describedfor tortuous contralateral femoral arteries in patientsundergoing TAVR (103).

The data on angiograms and PCI in patients withprior bypass surgery also favor the transfemoral route(104). More data are needed to support the routineuse of fluoroscopy, micropuncture needle with

ultrasound assistance, smaller sheath size, earlysheath removal, and use of the right dose of anti-coagulation agents that will likely lower bleedingand vascular complication rates from femoral access.We eagerly await the results of SAFARI-STEMI(Femoral Versus Radial Access for Primary PCI;NCT01398254) (105).

SUMMARY AND CONCLUSIONS

New iterations of catheters, closure devices, andanticoagulation doses and regimens have loweredbleeding rates following PCI; however, bleeding stillaccounts for significant morbidity and mortality.Tools to estimate an individual’s bleeding risk needto be routinely used. The bleeding risk is modifi-able; therefore, interventionalists are encouraged toadopt the safest and best practices to lower thebleeding rates. Operators need to choose the rightanticoagulation agent and administer the right dose.Switching to radial access for PCI will be a paradigmshift for operators who have previously usedfemoral access. It has a learning curve, and onlyhigh-volume operators will be able to demonstratelower bleeding rates and improved outcomes. Pro-ficiency in femoral access is still required. Adoptionof best practices to lower bleeding rates willcertainly improve quality of care and downstreamoutcomes of PCI.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Mandeep Singh, Division of Cardiovascular Diseases,Department of Internal Medicine, Mayo Clinic, 200First Street Southwest, Rochester, Minnesota 55905.E-mail: singh.mandeep@mayo.edu.

RE F E RENCE S

1. Amin AP, Marso SP, Rao SV, et al. Cost-effectiveness of targeting patients undergoingpercutaneous coronary intervention for therapywith bivalirudin versus heparin monotherapyaccording to predicted risk of bleeding. CircCardiovas Qual Outcomes 2010;3:358–65.

2. Eleid MF, Rihal CS, Gulati R, et al. Sys-tematic use of transradial PCI in patients withST-segment elevation myocardial infarction: acall to “arms.” J Am Coll Cardiol Intv 2013;6:1145–8.

3. Kwok CS, Rao SV, Myint PK, et al. Majorbleeding after percutaneous coronary interventionand risk of subsequent mortality: a systematicreview and meta-analysis. Open Heart 2014;1:e000021.

4. Mamas MA, Anderson SG, Carr M, et al., BritishCardiovascular Intervention Society and the Na-tional Institute for Cardiovascular Outcomes

Research. Baseline bleeding risk and arterial ac-cess site practice in relation to procedural out-comes after percutaneous coronary intervention.J Am Coll Cardiol 2014;64:1554–64.

5. Chhatriwalla AK, Amin AP, Kennedy KF, et al.,for the National Cardiovascular Data Registry.Association between bleeding events and in-hospital mortality after percutaneous coronaryintervention. JAMA 2013;309:1022–9.

6. Ndrepepa G, Schuster T, Hadamitzky M, et al.Validation of the Bleeding Academic ResearchConsortium definition of bleeding in patients withcoronary artery disease undergoing percutaneouscoronary intervention. Circulation 2012;125:1424–31.

7. Centers for Medicare &Medicaid Services.Details for title: Medicare Acute Care Episode(ACE) demonstration. 2009. Available at: http://www.cms.gov/Medicare/Demonstration-Projects/

DemoProjectsEvalRpts/Medicare-Demonstrations-Items/CMS1204388.html. Accessed April 1, 2015.

8. Marso SP, Amin AP, House JA, et al., for theNational Cardiovascular Data Registry. Associationbetween use of bleeding avoidance strategies andrisk of periprocedural bleeding among patientsundergoing percutaneous coronary intervention.JAMA 2010;303:2156–64.

9. Stone GW, Witzenbichler B, Guagliumi G, et al.,for the HORIZONS-AMI Trial Investigators. Biva-lirudin during primary PCI in acute myocardialinfarction. N Engl J Med 2008;358:2218–30.

10. Mehran R, Rao SV, Bhatt DL, et al. Standard-ized bleeding definitions for cardiovascular clinicaltrials: a consensus report from the Bleeding Aca-demic Research Consortium. Circulation 2011;123:2736–47.

11. Rao SV, McCoy LA, Spertus JA, et al. Anupdated bleeding model to predict the risk of

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2236

post-procedure bleeding among patients under-going percutaneous coronary intervention: areport using an expanded bleeding definition fromthe National Cardiovascular Data Registry CathPCIRegistry. J Am Coll Cardiol Intv 2013;6:897–904.

12. Bovill EG, Terrin ML, Stump DC, et al. Hemor-rhagic events during therapy with recombinanttissue-type plasminogen activator, heparin, andaspirin for acute myocardial infarction. Results ofthe Thrombolysis in Myocardial Infarction (TIMI),Phase II Trial. Ann Intern Med 1991;115:256–65.

13. Sabatine MS, Morrow DA, Giugliano RP, et al.Association of hemoglobin levels with clinicaloutcomes in acute coronary syndromes. Circula-tion 2005;111:2042–9.

14. Subherwal S, Peterson ED, Dai D, et al. Tem-poral trends in and factors associated withbleeding complications among patients undergo-ing percutaneous coronary intervention: a reportfrom the National Cardiovascular Data CathPCIRegistry. J Am Coll Cardiol 2012;59:1861–9.

15. Verheugt FW, Steinhubl SR, Hamon M, et al.Incidence, prognostic impact, and influence ofantithrombotic therapy on access and nonaccesssite bleeding in percutaneous coronary interven-tion. J Am Coll Cardiol Intv 2011;4:191–7.

16. Vavalle JP, Clare R, Chiswell K, et al. Prog-nostic significance of bleeding location andseverity among patients with acute coronary syn-dromes. J Am Coll Cardiol Intv 2013;6:709–17.

17. Doyle BJ, Ting HH, Bell MR, et al. Majorfemoral bleeding complications after percuta-neous coronary intervention: incidence, pre-dictors, and impact on long-term survival among17,901 patients treated at the Mayo Clinic from1994 to 2005. J Am Coll Cardiol Intv 2008;1:202–9.

18. Mehta SK, Frutkin AD, Lindsey JB, et al.Bleeding in patients undergoing percutaneouscoronary intervention: the development of a clin-ical risk algorithm from the National Cardiovas-cular Data Registry. Circ Cardiovasc Interv 2009;2:222–9.

19. Ahmed B, Piper WD, Malenka D, et al. Signifi-cantly improved vascular complications amongwomen undergoing percutaneous coronary inter-vention: a report from the Northern New EnglandPercutaneous Coronary Intervention Registry. CircCardiovasc Interv 2009;2:423–9.

20. Applegate RJ, Sacrinty MT, Kutcher MA, et al.Trends in vascular complications after diag-nostic cardiac catheterization and percutaneouscoronary intervention via the femoral artery, 1998to 2007. J Am Coll Cardiol Intv 2008;1:317–26.

21. Roe MT, Messenger JC, Weintraub WS, et al.Treatments, trends, and outcomes of acutemyocardial infarction and percutaneous coronaryintervention. J Am Coll Cardiol 2010;56:254–63.

22. Dauerman HL, Rao SV, Resnic FS, et al.Bleeding avoidance strategies: consensus andcontroversy. J Am Coll Cardiol 2011;58:1–10.

23. Rao SC, Chhatriwalla AK, Kennedy KF, et al.Pre-procedural estimate of individualized bleedingrisk impacts physicians’ utilization of bivalirudinduring percutaneous coronary intervention. J AmColl Cardiol 2013;61:1847–52.

24. Singh M, Lennon RJ, Gulati R, et al. Risk scoresfor 30-day mortality after percutaneous coronaryintervention: new insights into causes and risk ofdeath. Mayo Clin Proc 2014;89:631–7.

25. Singh M, Rihal CS, Lennon RJ, et al. Bedsideestimation of risk from percutaneous coronaryintervention: the new Mayo Clinic risk scores.Mayo Clin Proc 2007;82:701–8.

26. Singh M, Lennon RJ, Holmes DR Jr., et al.Correlates of procedural complications and asimple integer risk score for percutaneous coro-nary intervention. J Am Coll Cardiol 2002;40:387–93.

27. Mehran R, Pocock S, Nikolsky E, et al. Impact ofbleeding on mortality after percutaneous coronaryintervention results from a patient-level pooledanalysis of the REPLACE-2 (randomized evaluationof PCI linking angiomax to reduced clinical events),ACUITY (acute catheterization and urgent in-tervention triage strategy), and HORIZONS-AMI(harmonizing outcomes with revascularizationand stents in acute myocardial infarction) trials.J Am Coll Cardiol Intv 2011;4:654–64.

28. Daugherty SL, Thompson LE, Kim S, et al.Patterns of use and comparative effectiveness ofbleeding avoidance strategies in men and womenfollowing percutaneous coronary interventions: anobservational study from the National Cardiovas-cular Data Registry. J Am Coll Cardiol 2013;61:2070–8.

29. Kinnaird TD, Stabile E, Mintz GS, et al. Inci-dence, predictors, and prognostic implications ofbleeding and blood transfusion following percu-taneous coronary interventions. Am J Cardiol2003;92:930–5.

30. Feit F, VoeltzMD,AttubatoMJ, et al. Predictorsand impact of major hemorrhage on mortalityfollowing percutaneous coronary intervention fromthe REPLACE-2 Trial. Am J Cardiol 2007;100:1364–9.

31. Ndrepepa G, Berger PB, Mehilli J, et al. Peri-procedural bleeding and 1-year outcome afterpercutaneous coronary interventions: appropri-ateness of including bleeding as a component of aquadruple end point. J Am Coll Cardiol 2008;51:690–7.

32. Ndrepepa G, Schulz S, Keta D, et al. Bleedingafter percutaneous coronary intervention withBivalirudin or unfractionated Heparin and one-yearmortality. Am J Cardiol 2010;105:163–7.

33. Amlani S, Nadarajah T, Afzal R, et al. Mortalityand morbidity following a major bleed in a registrypopulation with acute ST elevation myocardialinfarction. J Thromb Thrombolysis 2010;30:434–40.

34. Doyle BJ, Rihal CS, Gastineau DA, et al.Bleeding, blood transfusion, and increased mor-tality after percutaneous coronary intervention:implications for contemporary practice. J Am CollCardiol 2009;53:2019–27.

35. Tsai AG, Cabrales P, Intaglietta M. Microvas-cular perfusion upon exchange transfusion withstored red blood cells in normovolemic anemicconditions. Transfusion 2004;44:1626–34.

36. Almac E, Ince C. The impact of storage on redcell function in blood transfusion. Best Pract ResClin Anaesthesiol 2007;21:195–208.

37. Reynolds JD, Ahearn GS, Angelo M, et al.S-nitrosohemoglobin deficiency: a mechanism forloss of physiological activity in banked blood. ProcNatl Acad Sci U S A 2007;104:17058–62.

38. Twomley KM, Rao SV, Becker RC. Pro-inflammatory, immunomodulating, and pro-thrombotic properties of anemia and red bloodcell transfusions. J Thromb Thrombolysis 2006;21:167–74.

39. Sherwood MW, Wang Y, Curtis JP, et al. Pat-terns and outcomes of red blood cell transfusionin patients undergoing percutaneous coronaryintervention. JAMA 2014;311:836–43.

40. Rao SV, Jollis JG, Harrington RA, et al. Rela-tionship of blood transfusion and clinical out-comes in patients with acute coronary syndromes.JAMA 2004;292:1555–62.

41. The EPIC Investigators. Use of a monoclonalantibody directed against the platelet glycopro-tein IIb/IIIa receptor in high-risk coronary angio-plasty. N Engl J Med 1994;330:956–61.

42. EPISTENT Investigators. Randomised placebo-controlled and balloon-angioplasty-controlledtrial to assess safety of coronary stenting withuse of platelet glycoprotein-IIb/IIIa blockade.Lancet 1998;352:87–92.

43. Cho L, Topol EJ, Balog C, et al. Clinical benefitof glycoprotein IIb/IIIa blockade with Abciximab isindependent of gender: pooled analysis from EPIC,EPILOG and EPISTENT trials. Evaluation of 7E3 forthe Prevention of Ischemic Complications. Evalu-ation in Percutaneous Transluminal Coronary An-gioplasty to Improve Long-Term Outcome withAbciximab GP IIb/IIIa blockade. Evaluation ofPlatelet IIb/IIIa Inhibitor for Stent. J Am Coll Car-diol 2000;36:381–6.

44. Alexander KP, Chen AY, Roe MT, et al., for theCRUSADE Investigators. Excess dosing of anti-platelet and antithrombin agents in the treatmentof non-ST-segment elevation acute coronarysyndromes. JAMA 2005;294:3108–16.

45. Tolleson TR, O’Shea JC, Bittl JA, et al. Rela-tionship between heparin anticoagulation andclinical outcomes in coronary stent intervention:observations from the ESPRIT trial. J Am CollCardiol 2003;41:386–93.

46. Levine GN, Bates ER, Blankenship JC, et al.2011 ACCF/AHA/SCAI guideline for percutaneouscoronary intervention: a report of the AmericanCollege of Cardiology Foundation/American HeartAssociation Task Force on Practice Guidelines andthe Society for Cardiovascular Angiography andInterventions. J Am Coll Cardiol 2011;58:e44–122.

47. Wallentin L, Becker RC, Budaj A, et al., PLATOInvestigators. Ticagrelor versus clopidogrel in pa-tients with acute coronary syndromes. N Engl JMed 2009;361:1045–57.

48. Mehta SR, Yusuf S, Peters RJ, et al., Clopi-dogrel in Unstable angina to prevent RecurrentEvents trial (CURE) Investigators. Effects of pre-treatment with clopidogrel and aspirin followedby long-term therapy in patients undergoingpercutaneous coronary intervention: the PCI-CUREstudy. Lancet 2001;358:527–33.

49. Mehta SR, Yusuf S, Clopidogrel in Unstableangina to prevent Recurrent Events (CURE) StudyInvestigators. The Clopidogrel in Unstable angina

J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5 SinghM A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8 Bleeding Complications Following PCI

2237

to prevent Recurrent Events (CURE) trial pro-gramme; rationale, design and baseline charac-teristics including a meta-analysis of the effects ofthienopyridines in vascular disease. Eur Heart J2000;21:2033–41.

50. Wiviott SD, Braunwald E, McCabe CH, et al.,for the TRITON–TIMI 38 Investigators. Prasugrelversus clopidogrel in patients with acute coronarysyndromes. N Engl J Med 2007;357:2001–15.

51. Bhatt DL, Lincoff AM, Gibson CM, et al., for theCHAMPION PLATFORM Investigators. Intravenousplatelet blockade with cangrelor during PCI.N Engl J Med 2009;361:2330–41.

52. Bhatt DL, Stone GW, Mahaffey KW, et al., forthe CHAMPION PHOENIX Investigators. Effect ofplatelet inhibition with cangrelor during PCI onischemic events. N Engl J Med 2013;368:1303–13.

53. Lincoff AM, Kleiman NS, Kereiakes DJ, et al.,for the REPLACE-2 Investigators. Long-term effi-cacy of bivalirudin and provisional glycoproteinIIb/IIIa blockade vs heparin and planned glyco-protein IIb/IIIa blockade during percutaneouscoronary revascularization: REPLACE-2 random-ized trial. JAMA 2004;292:696–703.

54. Lincoff AM, Bittl JA, Harrington RA, et al., forthe REPLACE-2 Investigators. Bivalirudin andprovisional glycoprotein IIb/IIIa blockade com-pared with heparin and planned glycoproteinIIb/IIIa blockade during percutaneous coronaryintervention: REPLACE-2 randomized trial. JAMA2003;289:853–63.

55. Stone GW, McLaurin BT, Cox DA, et al., for theACUITY Investigators. Bivalirudin for patients withacute coronary syndromes. N Engl J Med 2006;355:2203–16.

56. Kastrati A, Neumann FJ, Mehilli J, et al., forthe ISAR-REACT 3 Trial Investigators. Bivalirudinversus unfractionated heparin during percuta-neous coronary intervention. N Engl J Med 2008;359:688–96.

57. Steg PG, van ’t Hof A, Hamm CW, et al., for theEUROMAX Investigators. Bivalirudin started dur-ing emergency transport for primary PCI. N Engl JMed 2013;369:2207–17.

58. Cavender MA, Sabatine MS. Bivalirudin versusheparin in patients planned for percutaneouscoronary intervention: a meta-analysis of rando-mised controlled trials. Lancet 2014;384:599–606.

59. Shahzad A, Kemp I, Mars C, et al., for theHEAT-PPCI Trial Investigators. Unfractionatedheparin versus bivalirudin in primary percutaneouscoronary intervention (HEAT-PPCI): an open-label,single centre, randomised controlled trial. Lancet2014;384:1849–58.

60. Koutouzis M, Lagerqvist B, James S, et al.Unfractionated heparin administration in patientstreated with bivalirudin during primary percuta-neous coronary intervention is associated lowermortality and target lesion thrombosis: a reportfrom the Swedish Coronary Angiography and An-gioplasty Registry (SCAAR). Heart 2011;97:1484–8.

61. Han Y, Guo J, Zheng Y, et al., for the BRIGHTInvestigators. Bivalirudin vs Heparin With orWithout Tirofiban During Primary PercutaneousCoronary Intervention in Acute Myocardial

Infarction: The BRIGHT Randomized Clinical Trial.JAMA 2015 Mar 16 [E-pub ahead of print].

62. Valgimigli M. Bivalirudin infusion compared tounfractionated heparin in patients with acutecoronary syndromes undergoing invasive man-agement: results from the Minimizing AdverseHaemorrhagic Events By Transradial Access SiteAnd Systemic Implementation Of Angiox (matrix)Anti-thrombin Program. Paper presented at: ACC.15; March 15, 2015; San Diego, California.

63. Feldman DN, Swaminathan RV, Kaltenbach LA,et al. Adoption of radial access and comparison ofoutcomes to femoral access in percutaneous cor-onary intervention: an updated report from thenational cardiovascular data registry (2007–2012).Circulation 2013;127:2295–306.

64. Bangalore S, Bhatt DL. Femoral arterial accessand closure. Circulation 2011;124:e147–56.

65. Cantor WJ, Mahaffey KW, Huang Z, et al.Bleeding complications in patients with acutecoronary syndrome undergoing early invasivemanagement can be reduced with radial access,smaller sheath sizes, and timely sheath removal.Catheter Cardiovasc Interv 2007;69:73–83.

66. Ambrose JA, Lardizabal J, Mouanoutoua M,et al. Femoral micropuncture or routine introducerstudy (FEMORIS). Cardiology 2014;129:39–43.

67. Seto AH, Abu-Fadel MS, Sparling JM, et al.Real-time ultrasound guidance facilitates femoralarterial access and reduces vascular complications:FAUST (Femoral Arterial Access With UltrasoundTrial). J Am Coll Cardiol Intv 2010;3:751–8.

68. Bernat I, Horak D, Stasek J, et al. ST-segmentelevation myocardial infarction treated by radial orfemoral approach in a multicenter randomizedclinical trial: the STEMI-RADIAL trial. J Am CollCardiol 2014;63:964–72.

69. Romagnoli E, Biondi-Zoccai G, Sciahbasi A,et al. Radial versus femoral randomized investi-gation in ST-segment elevation acute coronarysyndrome: the RIFLE-STEACS (Radial VersusFemoral Randomized Investigation in ST-ElevationAcute Coronary Syndrome) study. J Am Coll Car-diol 2012;60:2481–9.

70. Jolly SS, Yusuf S, Cairns J, et al., for the RIVALtrial group. Radial versus femoral access for cor-onary angiography and intervention in patientswith acute coronary syndromes (RIVAL): a rando-mised, parallel group, multicentre trial. Lancet2011;377:1409–20.

71. Valgimigli M, Gagnor A, Calabro P, et al., forthe MATRIX Investigators. Radial versus femoralaccess in patients with acute coronary syndromesundergoing invasive management: a randomisedmulticentre trial. Lancet 2015 Mar 13 [E-pub aheadof print].

72. Bertrand OF, Jolly SS, Rao SV, et al. Meta-analysis comparing bivalirudin versus heparinmonotherapy on ischemic and bleeding outcomesafter percutaneous coronary intervention. Am JCardiol 2012;110:599–606.

73. Rao SV, Hess CN, Barham B, et al. A registry-based randomized trial comparing radial andfemoral approaches in women undergoing percu-taneous coronary intervention: the SAFE-PCI forWomen (Study of Access Site for Enhancement of

PCI for Women) trial. J Am Coll Cardiol Intv 2014;7:857–67.

74. Jolly SS, Cairns J, Yusuf S, et al., for the RIVALInvestigators. Procedural volume and outcomeswith radial or femoral access for coronary angi-ography and intervention. J Am Coll Cardiol 2014;63:954–63.

75. Dehmer GJ, Blankenship JC, Cilingiroglu M,et al. SCAI/ACC/AHA expert consensus Document:2014 update on percutaneous coronary interven-tion without on-site surgical backup. J Am CollCardiol 2014;63:2624–41.

76. Bundhoo S, Nallur-Shivu G, Ossei-Gerning N,et al. Switching from transfemoral to transradialaccess for PCI: a single-center learning curve over5 years. J Invasive Cardiol 2014;26:535–41.

77. Hess CN, Peterson ED, Neely ML, et al. Thelearning curve for transradial percutaneous coro-nary intervention among operators in the UnitedStates: a study from the National CardiovascularData Registry. Circulation 2014;129:2277–86.

78. Abbott JD. The pace of transradial procedurallearning. Circulation 2014;129:2250–2.

79. Kotowycz MA, Dzavik V. Radial artery patencyafter transradial catheterization. Circ CardiovascInterv 2012;5:127–33.

80. Pancholy S, Coppola J, Patel T, et al. Pre-vention of radial artery occlusion-patent hemo-stasis evaluation trial (PROPHET study): arandomized comparison of traditional versuspatency documented hemostasis after transradialcatheterization. Catheter Cardiovasc Interv 2008;72:335–40.

81. Cubero JM, Lombardo J, Pedrosa C, et al.Radial compression guided by mean artery pres-sure versus standard compression with a pneu-matic device (RACOMAP). Catheter CardiovascInterv 2009;73:467–72.

82. Rao SV, Tremmel JA, Gilchrist IC, et al. Bestpractices for transradial angiography and inter-vention: a consensus statement from the societyfor cardiovascular angiography and intervention’stransradial working group. Catheter CardiovascInterv 2014;83:228–36.

83. Valgimigli M, Campo G, Penzo C, et al., for theRADAR Investigators. Transradial coronary cathe-terization and intervention across the wholespectrum of Allen test results. J Am Coll Cardiol2014;63:1833–41.

84. Koreny M, Riedmuller E, Nikfardjam M, et al.for the Arterial puncture closing devices comparedwith standard manual compression after cardiaccatheterization: systematic review and meta-analysis. JAMA 2004;291:350–7.

85. Hvelplund A, Jeger R, Osterwalder R, et al.The Angio-Seal femoral closure device allows im-mediate ambulation after coronary angiographyand percutaneous coronary intervention. Euro-Intervention 2011;7:234–41.

86. Fram DB, Giri S, Jamil G, et al. Suture closureof the femoral arteriotomy following invasivecardiac procedures: a detailed analysis of efficacy,complications, and the impact of early ambulationin 1,200 consecutive, unselected cases. CatheterCardiovasc Interv 2001;53:163–73.

Singh J A C C V O L . 6 5 , N O . 2 0 , 2 0 1 5

Bleeding Complications Following PCI M A Y 2 6 , 2 0 1 5 : 2 2 2 5 – 3 8

2238

87. SanbornTA, TomeyMI,MehranR, et al. Femoralvascular closure device use, bivalirudin anti-coagulation, and bleeding after primary angio-plasty for STEMI: results from the HORIZONS-AMItrial. Catheter Cardiovasc Interv 2015;85:371–9.

88. Hermanides RS, Ottervanger JP, Dambrink JH,et al., Zwolle Myocardial Infarction study Group.Closure device or manual compression in patientsundergoing percutaneous coronary intervention: arandomized comparison. J Invasive Cardiol 2010;22:562–6.

89. Sanborn TA, Ebrahimi R, Manoukian SV, et al.Impact of femoral vascular closure devices andantithrombotic therapy on access site bleeding inacute coronary syndromes: The Acute Catheteri-zation and Urgent Intervention Triage Strategy(ACUITY) trial. Circ Cardiovasc Interv 2010;3:57–62.

90. Gurm HS, Hosman C, Share D, et al. Compar-ative safety of vascular closure devices andmanual closure among patients having percuta-neous coronary intervention. Ann Intern Med2013;159:660–6.

91. Tavris DR, Wang Y, Jacobs S, et al. Bleedingand vascular complications at the femoral accesssite following percutaneous coronary intervention(PCI): an evaluation of hemostasis strategies.J Invasive Cardiol 2012;24:328–34.

92. Biancari F, D’Andrea V, Di Marco C, et al. Meta-analysis of randomized trials on the efficacy ofvascular closure devices after diagnostic angiog-raphy and angioplasty. AmHeart J 2010;159:518–31.

93. Resnic FS, Gross TP, Marinac-Dabic D, et al.Automated surveillance to detect postproceduresafety signals of approved cardiovascular devices.JAMA 2010;304:2019–27.

94. Resnic FS, Wang TY, Arora N, et al. Quan-tifying the learning curve in the use of a novel

vascular closure device: an analysis of theNCDR (National Cardiovascular Data Registry)CathPCI registry. J Am Coll Cardiol Intv 2012;5:82–9.

95. Nehler MR, Lawrence WA, Whitehill TA, et al.Iatrogenic vascular injuries from percutaneousvascular suturing devices. J Vasc Surg 2001;33:943–7.

96. Bangalore S, Arora N, Resnic FS. Vascularclosure device failure: frequency and implications:a propensity-matched analysis. Circ CardiovascInterv 2009;2:549–56.

97. Vidi VD, Matheny ME, Govindarajulu US,et al. Vascular closure device failure in contem-porary practice. J Am Coll Cardiol Intv 2012;5:837–44.

98. Tsai TT, Patel UD, Chang TI, et al. Contem-porary incidence, predictors, and outcomes ofacute kidney injury in patients undergoing percu-taneous coronary interventions: insights from theNCDR Cath-PCI registry. J Am Coll Cardiol Intv2014;7:1–9.

99. Stone GW, Witzenbichler B, Guagliumi G,et al., HORIZONS-AMI Trial Investigators. Heparinplus a glycoprotein IIb/IIIa inhibitor versus biva-lirudin monotherapy and paclitaxel-eluting stentsversus bare-metal stents in acute myocardialinfarction (HORIZONS-AMI): final 3-year resultsfrom a multicentre, randomised controlled trial.Lancet 2011;377:2193–204.

100. Hamon M, Pristipino C, Di Mario C, et al.Consensus document on the radial approachin percutaneous cardiovascular interventions: po-sition paper by the European Association ofPercutaneous Cardiovascular Interventions andWorking Groups on Acute Cardiac Care** andThrombosis of the European Society of Cardiology.EuroIntervention 2013;8:1242–51.

101. Jolly SS, Amlani S, Hamon M, et al. Radialversus femoral access for coronary angiography orintervention and the impact on major bleeding andischemic events: a systematic review and meta-analysis of randomized trials. Am Heart J 2009;157:132–40.

102. Bernelli C, Chieffo A, Montorfano M, et al.Usefulness of baseline activated clotting time-guided heparin administration in reducing bl-eeding events during transfemoral transcatheteraortic valve implantation. J Am Coll Cardiol Intv2014;7:140–51.

103. Curran H, Chieffo A, Buchanan GL, et al.A comparison of the femoral and radial crossovertechniques for vascular access management in trans-catheter aortic valve implantation: the Milan experi-ence. Catheter Cardiovasc Interv 2014;83:156–61.

104. Michael TT, Alomar M, Papayannis A, et al.A randomized comparison of the transradial andtransfemoral approaches for coronary arterybypass graft angiography and intervention: theRADIAL-CABG Trial (RADIAL Versus Femoral Ac-cess for Coronary Artery Bypass Graft Angiographyand Intervention). J Am Coll Cardiol Intv 2013;6:1138–44.

105. Ottawa Heart Institute Research Corpora-tion. Femoral Versus Radial Access for PrimaryPC (SAFARI-STEMI). 2015. Available at: https://clinicaltrials.gov/ct2/show/NCT01398254. AccessedApril 2, 2015.

106. Schomig A. Ticagrelor—is there need for anew player in the antiplatelet-therapy field?N Engl J Med 2009;361:1108–11.

KEY WORDS anticoagulants, plateletaggregation inhibitors, thrombosis,vascular closure devices