Clinical Impact of Bleeding
Risk Factors for Bleeding on Antiplatelet Therapy
Bleeding under antiplatelet therapy (APT) can occur in the early phase postprocedure
(e.g., after coronary or valve interventions), in the intermediate phase from discharge
during the first year, or in the chronic phase after the first year of treatment.
Major and minor risk factors for bleeding under APT have been recently issued by the
Academic Research Consortium. A high bleeding risk (HBR) status is defined by the
presence of at least one major or two minor risk criteria ([Fig. 1]).[1]
Fig. 1 Major and minor criteria for definition of high bleeding risk, modified according
to the Academic Research Consortium.[1] High bleeding risk is defined by the presence of at least one major or two minor
criteria.
Incidence and Case Fatality of Bleeding on Antiplatelet Therapy
Bleeding is associated with higher short-term, i.e., intrahospital, intermediate (up
to 12 months), and long-term (beyond 12 months) mortality after the event, which is
either caused by case fatality of the bleeding event itself or by thromboembolic complications
after temporary or permanent discontinuation/disruption or modification of APT.[2]
[3]
[4]
[5]
[6]
[7] In many cases, treating physicians tend to discontinue APT even after nonmajor bleeding
events. Cessation of antithrombotic therapy has been defined in the literature as
discontinuation (physician-recommended withdrawal), interruption (temporary cessation of antiplatelet treatment due to surgical necessity with resumption
within 14 days), or disruption (cessation of antiplatelet treatment due to bleeding or noncompliance).[8]
[9] In the Patterns of nonadherence to Antiplatelet Regimens In Stented patients (PARIS)
registry,[8] the risk of major adverse cardiovascular events (MACE) was approximately sevenfold
higher in the first week after disruption of dual antiplatelet therapy (DAPT). An
observational cohort study suggested that those who discontinued P2Y12 inhibitor after
3 months or had poor initial adherence had higher MACE risk compared with those with
persistently good adherence over 12 months.[10]
If a patient has a bleeding tendency, the majority of major bleeding events occur
early (within 3 months) after initiation of DAPT. Shortening of DAPT duration (from
3 to 1 month) is associated with a significant decrease of major bleeding in patients
with a HBR (2.1 vs. 1.0%).[11] Although the rates of major bleeding decrease over time, there remains a persistent
annual major bleeding rate of approximately 0.4% with low-dose acetyl salicylic acid
(ASA) and 0.6 to 0.9% with DAPT in the chronic phase beyond 1 year.[12] Among patients treated with DAPT, gastrointestinal bleeding (GIB) accounts for about
two-thirds of all bleeding events.[13] GIB occurs in 3 to 4% of cases in the first year of DAPT.[13] The bleeding rates reported in randomized trials need to be evaluated with caution
when translating into the real world. In clinical trials, patients often either underwent
a run-in period of event-free DAPT to be included or were excluded due to bleeding
risk criteria ([Table 1]). Precise numbers of major and clinically relevant nonmajor bleeding events in true
all-comer populations are sparse but likely higher than in well-controlled clinical
trials.[14] In addition, bleeding rates differ with regard to bleeding definition, type of APT,
and indication, timing and duration of APT ([Table 1]). In the PLATelet inhibition and patient Outcomes (PLATO) trial, non-coronary artery
bypass graft (non-CABG)-related major bleeding occurred at similar rates under ticagrelor
and clopidogrel (2.47 vs. 2.21%) during the first 30 days, and at a significantly
higher rate of 2.17% versus 1.65% after 30 days until the end of study treatment with
ticagrelor.[15] Similarly, in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing
Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction (TRITON)-Thrombolysis
In Myocardial Infarction (TIMI) 38 trial, TIMI major non-CABG-related bleeding rates
with prasugrel were similar to clopidogrel during the first 3 days (0.74 vs. 0.61%)
but were significantly higher with the use of prasugrel from 3 days to the end of
study drug treatment (1.71 vs. 1.23%).[16] In the Prevention of Cardiovascular Events in Patients with Prior Heart Attack Using
Ticagrelor Compared with Placebo on a Background of Aspirin–Thrombolysis in Myocardial
Infarction (PEGASUS)-TIMI 54 trial, long-term DAPT with ticagrelor beyond 1 year post-myocardial
infarction (MI) was associated with a 3-year major bleeding rate of 2.6% (90 mg bid)
and 2.3% (60 mg bid) versus 1.1% with ASA alone and intracranial hemorrhage (ICH)
rates of 0.56% (90 mg bid) and 0.61% (60 mg bid) versus 0.47% with ASA alone.[17] Similarly, moderate-to-severe bleeding events occurred in 2.0% of cases under long-term
treatment with clopidogrel plus ASA in the DAPT trial. Case fatality was high with
an annualized mortality rate after a bleeding event of 21.5 (95% confidence interval
[CI): 15.4–29.1) per 100 person-years.[18]
Table 1
Risk and fatality of major bleeding events in clinical trials and large observational
studies of oral antiplatelet therapy
|
Type of APT
|
Indication
|
Incidence/risk increase of major bleeding events
|
Incidence of intracranial hemorrhage/fatal bleedings
|
|
SAPT
|
|
Low-dose ASA
|
Primary prevention
|
0.23 per 100 patient-years[86];
29% increased risk of major bleeding (mostly gastrointestinal) in diabetics,[35] 87% increased risk of upper GIB in healthy elderly,[36] 37% increased risk for ICB versus control[87]
|
0.3%, no increase versus control[36]
|
|
Secondary prevention in CVD patients
|
Major bleeding 1.9%[88]; ∼60% increased risk of major extracranial bleeds (mostly gastrointestinal) compared
with placebo[89]
[90]
[91]
|
FB: 0.1%[88]
|
|
Clopidogrel monotherapy
|
Secondary prevention in patients with a history of ischemic stroke, myocardial infarction,
or
symptomatic atherosclerotic peripheral arterial disease.
|
Severe bleeding with clopidogrel versus ASA monotherapy: 1.38 versus 1.55% (nonsignificant);
severe GIB (0.49 vs. 0.71%, p < 0.05)[38]
|
Nonfatal primary ICH or FB: 0.39% in the clopidogrel group versus 0.53% in the aspirin
group[38]
|
|
Secondary prevention after PCI (event free under DAPT for 6–18 months)
|
1.2% over 2 years (vs. 2.0% compared with ASA monotherapy)[39]
|
ICH: 0.2% over 2 years
FB: not reported separately
|
|
Ticagrelor monotherapy
|
PCI with high-risk criteria (event free under DAPT for 3 months)
|
BARC bleeding 3–5 1.0% over 12 months[41]
BARC 3 or 5 bleeding 1.6% over 12 months in HBR patients[82]
|
ICH/FB: not reported separately
|
|
DAPT
|
|
ASA + clopidogrel
|
ACS ± PCI
|
Non-CABG PLATO-major bleeding 3.8%, non-CABG TIMI-major bleeding 2.2%[15]
|
FB: 0.3%[15]
|
|
Chronic coronary syndromes
|
Major bleeding 0.2 to 1.3% with short duration and 0.2 to 2.3% with long duration
DAPT[92]
|
|
|
Secondary prevention in patients with documented CAD, cerebrovascular disease, PAD,
or multiple atherothrombotic risk factors
|
GUSTO severe bleeding according 1.7% in the ASA + clopidogrel group and 1.3% in the
ASA + placebo group (relative risk: 1.25; 95 CI: 0.97–1.61; p = 0.09)[93]
|
Primary ICH 0.3 in both ASA + clopidogrel and ASA + placebo-treated patients
FB: 0.3 (ASA + clopidogrel) versus 0.2% (ASA + placebo)[93]
|
|
Extended therapy
|
BARC 3 bleeding 2.6% during 18 months extended treatment[94]
|
FB: 0.1%, no increase versus placebo[18]
|
|
ASA + prasugrel
|
ACS + PCI
|
Non-CABG TIMI-major bleeding 2.4%[95]
|
FB: 0.4%, increased versus ASA + clopidogrel (HR: 4.19)[95]
|
|
ASA + ticagrelor 90 mg/bid
|
ACS ± PCI
|
Non-CABG major bleeding 4.5%[15]
|
FB: 0.3%, no increase versus ASA + clopidogrel[15]
|
|
ASA + ticagrelor 60 mg/bid
|
Extended therapy (beyond 12 months, post-MI)
|
Year 1: 0.86%; year 2: 0.86%; year 3: 0.6%[12]
|
ICH: year 1: 0.15%; year 2: 0.18%; year 3: 0.28%
FB: year 1: 0.11%; year 2: 0.04%; year 3: 0.11%[12]
|
Abbreviations: ACS, acute coronary syndrome; ASA, acetyl salicylic acid; BARC, Bleeding
Academic Research Consortium; CABG, coronary artery bypass graft; CAD, coronary artery
disease; CI, confidence interval; CVD, cardiovascular disease; FB, fatal bleeding;
HR, hazard ratio; ICH, intracranial hemorrhage; ISTH, International Society on Thrombosis
and Haemostasis; MI, myocardial infarction; PCI, percutaneous coronary intervention;
TIMI, thrombolysis in myocardial infarction; UGIB, upper gastrointestinal bleeding.
Trade-Off between Bleeding and Ischemic Risk
Clear classification of the bleeding severity according to established bleeding risk
classifications (i.e., Bleeding Academic Research Consortium [BARC], International
Society of Thrombosis and Haemostasis [ISTH] bleeding classification; [Table 2]) and the individual assessment of bleeding recurrence and thrombotic risk are of
utmost importance. In a retrospective study of the Thrombin Receptor Antagonist for
Clinical Event Reduction in Acute Coronary Syndrome (TRACER) trial, the risk for mortality
after major bleeding (BARC 3b + c) was almost equal to the mortality risk after MI.
However, compared with bleeding with lower severity (BARC 2 and 3a), the mortality
after MI clearly outweighed the bleeding-associated mortality risk.[19] Thus, the trade-off between cardiovascular versus bleeding-associated mortality
needs to be clearly defined ([Fig. 2]). Although scores that weigh the antithrombotic benefit against the bleeding risk
of continued DAPT have not been evaluated up to now in acute bleeding situations,
they might aid decision-making about the net benefit of prolonged APT in HBR patients.
Several contemporary scores evaluate the bleeding risk and trade-off between bleeding
and ischemic risk, and weigh the net benefit of prolonged versus shortened DAPT. Their
utility in clinical routine is limited by their inability to discriminate between
bleeding and ischemic risks as well as by the characteristics of the patient population
and APT in the derivation cohorts[1]
[20]
[21]
[22]
[23]
[24]
[25] ([Table 3]).
Fig. 2 Weighing of ischemic versus bleeding risk (figure was composed by using Adobe Stock
vectors).
Table 2
Definition of major (non-fatal, non-surgical) and clinically-relevant non-major bleeding
events, according to ISTH and BARC definitions
|
ISTH96,97
|
BARC98
|
|
Major bleeding
|
• Fatal bleeding, and/or
• Symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal,
intraocular, retroperitoneal, intra-articular or pericardial, or intramuscular with
compartment syndrome, and/or
• Bleeding causing a fall in hemoglobin level of 2 g/dL or more or leading to transfusion
of two or more units of whole blood or red cells.
|
Type 3a
• Overt bleeding plus hemoglobin drop of 3 to 5 g/dL (provided hemoglobin drop is
related to bleed)
• Any transfusion with overt bleeding
Type 3b
• Overt bleeding plus hemoglobin drop of 5 g/dL (provided hemoglobin drop is related
to bleed)
• Cardiac tamponade
• Bleeding requiring surgical intervention for control (excluding dental/nasal/skin/hemorrhoid)
• Bleeding requiring intravenous vasoactive agents
Type 3c
• Intracranial hemorrhage (does not include microbleeds or hemorrhagic transformation,
does include intraspinal)
• Subcategories confirmed by autopsy or imaging or lumbar puncture
• Intraocular bleed compromising vision
|
|
Clinically relevant nonmajor bleeding
|
Any sign or symptom of hemorrhage (e.g., more bleeding than would be expected for
a clinical circumstance, including bleeding found by imaging alone) that does not
fit the criteria for the ISTH definition of major bleeding but does meet at least
one of the following criteria:
• requiring medical intervention by a health care professional
• leading to hospitalization or increased level of care
• prompting a face-to-face (i.e., not just a telephone or electronic communication)
evaluation
|
Type 2
Any overt, actionable sign of hemorrhage (e.g., more bleeding than would be expected
for a clinical circumstance, including bleeding found by imaging alone) that does
not fit the criteria for type 3, 4, or 5 but does meet at least one of the following
criteria:
• requiring nonsurgical, medical intervention by a health care professional,
• leading to hospitalization or increased level of care, or
• prompting evaluation.
|
Abbreviations: BARC, Bleeding Academic Research Consortium; ISTH, International Society
on Thrombosis and Haemostasis.
Table 3
Contemporary risk scores to define bleeding risk and trade-off between bleeding and
ischemic risk in patients treated with APT
|
Score
|
Patient population/predominant type of APT
|
Risk factors in the primary model
|
Correlation with bleeding in the derivation cohort
|
Trade-off between bleeding and ischemic risk
|
Clinical utility/validity in real world
|
|
DAPT (total score range: −2 to 10)[22]
|
PCI patients completing 12-month DAPT (60% ASA + clopidogrel, 40% ASA + prasugrel)
|
Age (categorized)
Smoking
Diabetes mellitus
MI at presentation
Prior PCI or MI
Paclitaxel-eluting stent
Stent diameter < 3 mm
CHF or LVEF < 30
Vein graft stent
|
AUC 0.68 for moderate to severe bleedings (GUSTO) between 12 and 30 months under of
prolonged DAPT.
|
Score ≥2 is associated with a
favorable benefit/risk ratio for prolonged DAPT, whereas a
score <2 is associated with an unfavorable benefit/risk ratio
|
Ability to stratify ischemic benefit across different DAPT patient cohorts[99]
[100]
[101]
[102]
|
|
PARIS (range: 0–14)[23]
|
PCI + drug-eluting
stent (>90% ASA + clopidogrel)
|
Score for bleeding (PARIS MB): age (percentiles > 50 years)
BMI
Current smoking
Anemia
CrCl < 60 mL/min
Triple therapy on discharge
|
AUC for major bleeding (BARC 3 or 5) at 2 years: 0.72
|
|
AUC 0.64 validation cohort (ADAPT-DES)[103]
|
|
PRECISE-DAPT (range: 0–100)[20]
|
PCI + stent (ASA + clopidogrel (88%))
|
Age (continuous), creatinine clearance, hemoglobin, previous bleeding, and white blood
cell count (simplified version excluding white blood cell count[104]
|
AUC 0.71 for TIMI major bleedings within 12 months
|
Longer DAPT duration reduced the composite ischemic endpoint (MI, definite ST, stroke,
or target vessel revascularization) in those at non-high bleeding risk
|
Recommended in current guidelines,[7]
[31] reproducible in real-world cohorts, limited usefulness in patients with MI and pre-existing
risk factors for bleeding[103]
[105]
[106]
|
|
PRAISE[24]
|
ACS (ASA + clopidogrel (64%), + ticagrelor (21%), + prasugrel (15%))
|
16 clinical variables[a], five therapeutic variables[b], two angiographic variables[c], and two procedural variables[d]
|
AUC for major bleeding within 12 months: 0.70 in the internal validation cohort
|
Robust hypothetical trade-offs in the occurrence of ischemic and bleeding events according
to individual MI and major bleeding score classes[24]
|
High predictability and good discrimination of the trade-off (AUC >0.8 for mortality,
MI, and major bleeding in an external validation cohort).[24] Allowing for differentiation of the use of potent P2Y12 inhibitors in different
bleeding/thrombotic risk subsets.[107] Further external validation required.
|
|
ARC-High Bleeding Risk Trade-off Model[1]
[21]
|
Consensus definition of HBR in clinical trials evaluating the safety and effectiveness
of devices and drug regimens for patients undergoing PCI
|
Combination of major and minor risk factors; see [Fig. 1]
|
No AUC in primary publication/consensus document
|
Eight independent predictors of MI and/or ST, 8 predictors for risk of BARC types
3 to 5 bleeding; both risk models showed moderate discrimination; AUC = 0.69 for predicting
MI and/or ST and 0.68 for predicting BARC types 3 to 5 bleeding.[21]
|
Recommended in current guidelines,[7]
[31] reproducible validation in external cohorts, high sensitivity, limited specifity[108]
[109]
|
|
BleeMACS (range: 0–80)[25]
|
Patients with ACS undergoing PCI (ASA + clopidogrel (91%), + ticagrelor (4%), + prasugrel
(5%))
|
Age (categories),
Hypertension, vascular disease, history of bleeding,
Malignancy, creatinine (categories) hemoglobin
|
AUC: 0.71 for 1-year major bleeding in derivation cohort (0.72 internal validation)
AUC: 0.65 (ACS + PCI);
AUC: 0.63 (ACS)
|
Not validated
|
Acceptable performance in elderly ACS patients[110]; further validation required
|
Abbreviations: ACS, acute coronary syndrome; ASA, acetyl salicylic acid; AUC, area
under the curve; BARC, Bleeding Academic Research Consortium; CHF, congestive heart
failure; DAPT, dual antiplatelet therapy; LVEF, left ventricular ejection fraction;
MI, myocardial infarction; PCI, percutaneous coronary intervention; TIMI, thrombolysis
in myocardial infarction.
a Age, sex, diabetes, hypertension, hyperlipidemia, peripheral artery disease, estimated
glomerular filtration rate (EGFR; using the Modification of Diet in Renal Disease
study formula), previous myocardial infarction, previous percutaneous coronary intervention,
previous coronary artery bypass graft, previous stroke, previous bleeding, malignancy,
STEMI at presentation, hemoglobin, and left ventricular ejection fraction [LVEF].
b Treatment with β blockers, angiotensin-converting enzyme inhibitors or angiotensin-receptor
blockers, statins, oral anticoagulation, and proton-pump inhibitors.
c Multivessel disease and complete revascularization.
d Vascular access and percutaneous coronary intervention with drug-eluting stent.
Evaluation before Resumption of Antiplatelet Therapy
Has the Bleeding Stopped?
Determining whether a major bleeding event has stopped depends on the source and severity
of the bleeding and may require multidisciplinary evaluation. Repeated diagnostic
assessment might be required to determine cessation of a bleeding event, including
close measurements (daily or even hourly depending on urgency of bleeding) to document
hemoglobin drops. Repeated computed tomography (CT) scans can help to determine the
time point of cessation of ICH and complications of the bleeding potentially requiring
neurosurgical intervention (e.g., hydrocephalus). In addition, CT angiography and
post-contrast CT may aid in identifying patients at high risk of hematoma expansion
according to the presence of contrast within the bleed, defined as “spot signs,” or
brain vascular malformation.[26] Scores that included further information on number of spot signs, dimension, and
attenuation have been shown to be predictive for mortality and poor outcome.[27] Evaluation of spot sign dynamics on CT perfusion source images might yield even
better prognostic information with regard to hematoma expansion.[28]
In case of GIB, repeated endoscopy may provide information about the localization
and cause of the bleeding. The use of endoscopy can effectively contribute to treatment
of the bleeding lesion, e.g., by epinephrine injection, cauterization, variceal band
ligation, and metal clip placement. Endovascular angiography[29] and medical options such as octreotide infusions may also be considered for bleeding
from angioectasia,[30] when endoscopic therapies fail or when bleeding originates from less accessible
locations such as the small bowel. Thus, in many situations, GIB can be effectively
controlled to allow a timely resumption of APT if indicated.
Is Antiplatelet Therapy Still Needed?
The question whether APT is still needed after a bleeding event and, if so, at what
potency and duration is subject to individual risk–benefit assessment. In many cases,
antithrombotic therapy can be shortened according to recent guideline recommendations.[31] Individualized strategies to reduce the bleeding risk by mitigating APT should include
the type of intervention and the initial drug strategy. Thus, use of an abbreviated
DAPT duration with ultrathin stents has been shown to reduce bleeding risk in HBR
patients without substantially increasing ischemic risk, including stent thrombosis.[32] Recently, recommendations to guide the design of clinical trials of devices and
drugs in HBR patients undergoing percutaneous coronary intervention (PCI) have been
proposed.[33]
ASA Used for Primary Prevention in Individuals with Low or Moderate Risk for Ischemic
Events
ASA may be considered for primary prevention in high- and very high-risk individuals
according to the SCORE2 and SCORE2-OP risk classifications[34] given its unfavorable net benefit versus placebo in moderate risk or diabetic patients.[35]
[36] Resumption of ASA used for primary prevention after a bleeding event should therefore
be avoided in low-risk patients and discussed on a case-by-case approach in high-
and very high-risk patients given its accumulating benefit over time.[37]
SAPT for Secondary Prevention
The most common single antiplatelet therapy (SAPT) for secondary prevention in cardio-,
cerebrovascular, and peripheral artery disease is aspirin. However, there are cumulative
data demonstrating clopidogrel monotherapy to be as effective and safer than aspirin
for chronic maintenance monotherapy.[38]
[39]
Dual Antiplatelet Therapy
Duration of DAPT can be reduced down to 1 month in patients undergoing elective PCI
with modern stents and to 1 to 3 months after acute coronary syndrome (ACS) in high
and very high bleeding risk patients according to recent guidelines.[31]
[32]
[40] De-escalation to less potent P2Y12 inhibitors or early switching to P2Y12 inhibitor monotherapy has been established as effective strategies to prevent bleeding
in HBR patients and in patients in whom the trade-off prevails on the bleeding risk
side ([Table 3]).[41]
[42]
[43]
[44]
[45]
[46]
What Caused the Bleeding?
Risk factors for bleeding can be classified into reversible and irreversible factors
([Table 4]). The question of what caused the bleeding is closely linked to the question of
which modifiable factors may have contributed to the bleed and whether definitive
treatment of these factors has been achieved. Identification of reversible/modifiable
and irreversible factors as well as the risk for recurrence is important to provide
an individual strategy to resume APT.
Table 4
Risk of major bleeding recurrence in clinical trials and observational studies of
antiplatelet therapy
|
Reversible causes of bleeding
|
Irreversible causes of bleeding
|
|
General factors
|
|
Arterial hypertension
Alcohol abuse
Drug abuse
|
Older age (≥75 years)
History of major spontaneous bleeding
Malignancy
Genetic factors
Cirrhotic liver disease
Uncontrollable arterial hypertension
Hematologic abnormalities (e.g., low platelet count <50.000 per µL)
Dialysis-dependent kidney disease or renal transplantation[111]
|
|
Specific factors
|
|
GI bleeding
|
|
Small bleeding ulcer that can be treated by endoscopic therapy
Bleeding from gastric ulcer associated with Helicobacter disease
|
Bleeding from sources that cannot be sufficiently treated by surgery or endoscopic
procedures e.g., colonic tumor, angiodysplasia
|
|
Intracranial bleeding
|
|
Traumatic cause of ICH
Poor INR control, overdosing of antithrombotic therapies
|
Cerebral amyloid angiopathy
Vascular abnormalities that cannot be treated by surgical or interventional procedures.
Spontaneous ICH without identifiable risk factors (e.g., poor INR control)
|
Abbreviations: ICH, intracranial hemorrhage; INR, international normalized ratio.
Recurrence of Bleeding after Resumption of Antiplatelet Therapy
The dilemma of weighing the benefit against the harm of resumption of antithrombotic
therapy warrants interdisciplinary discussion for defining both severity of bleeding
and risks of bleeding recurrence and thromboembolic events. [Table 4] summarizes risk factors for bleeding recurrence after ICH and major GIB. In addition,
bleeding event history and comorbidities add to the bleeding risk recurrence and should
be carefully included in the risk assessment, especially if risk factors cannot be
controlled. Additional bleeding risk factors include those listed in the ARC-HBR classification
scheme ([Fig. 1]).[1] The use of biomarkers like growth differentiation factor-15 has been investigated
to determine recurrent bleeding risk and recurrent events in post-hoc analyses of
large clinical trials.[47] However, its use in clinical routine is still limited.
Risk of Recurrence of Gastrointestinal Bleeding
Predictors of recurrent GIB risk and bleeding fatality depend on the cause of bleeding.
Thus, worse outcomes are associated with chronic alcoholism and active cancer.[22] Risk scoring systems have been evaluated to predict the risk of mortality and rebleeding
after GIB. The Rockall scoring system[48] and the Glasgow–Blatchford Scale[49] are scoring systems that include clinical, laboratory, and/or endoscopic findings.
These risk scores are used to predict the mortality associated with GIB, the urgency
of endoscopy, and likelihood of endoscopic intervention, but they have not been validated
for guidance regarding resumption of APT. The endoscopic appearances of a peptic ulcer,
as described by the Forrest classification,[50] may be used to predict the risk of rebleeding in the period soon after an endoscopy.
Patients with actively bleeding ulcers or ulcers with a visible blood vessel have
a much greater risk of rebleeding than ulcers with a clean base.[51] It is important to evaluate whether the GIB occurred during proton-pump inhibitor
(PPI) treatment or not. If the upper GIB happens under effective PPI treatment, this
might be an indicator to stop or reduce APT on an individual basis if there are no
other reversible factors (Helicobacter pylori [HP], nonsteroidal anti-inflammatory drug, endoscopic therapy). Initiation of effective
gastroprotection is a way to prevent recurrent bleeding. A randomized placebo-controlled
trial showed that high-dose intravenous PPI treatment (omeprazole 80 mg bolus then
8 mg/h infusion for 72 hours after endoscopic therapy)[52] in patients who had undergone endoscopic treatment of peptic ulcers effectively
reduced the risk of rebleeding compared with endoscopy and placebo. Recent meta-analyses
support the use of PPI therapy to promote healing of peptic ulcers thus reducing bleeding
and mortality in patients at risk for upper GIB.[53]
[54] There is no randomized controlled trial (RCT) that has evaluated the use of different
PPI regimens and durations in patients that require resumption of antithrombotic therapy
after major GIB. The common-sense view of the authors is that PPI therapy should be
continued indefinitely at standard dose in patients with major upper GIB unless there
are clear contraindications. Active HP infection is frequent with a reported prevalence
of 20% in acute MI patients and still is an underestimated risk factor for GIB in
patients treated with DAPT.[55]
[56]
[57] Therefore, HP screening, eradication, and gastroprotective measures are important
to minimize recurrent GIB risk.[57]
[58]
[59]
Risk of Recurrence after Intracranial Bleeding
The REstart or STop Antithrombotics Randomized Trial (RESTART) evaluated the efficacy
and safety of restarting APT after ICH. APT could be restarted in the interventional
arm by using either one or a combination of ASA, dipyridamole, or clopidogrel at any
time beyond 24 hours (median: 76 days) after randomization. In the comparator arm,
no APT was allowed. However, it was permitted to start or discontinue APT or anticoagulant
therapy if clinically indicated during follow-up, regardless of initial treatment
allocation. Interestingly, RESTART demonstrated a reduced risk of both thromboembolic
events and hemorrhagic strokes in patients restarting APT.[60]
[61] These observations were recently confirmed in an extended follow-up at a median
of 3.0 years.[62] Given the variable time points of restarting APT in RESTART, the optimal time interval
from ICH is still uncertain.
Arterial hypertension is a major cause underlying ICH, and consensus exists that blood
pressure should be well controlled after ICH.[63] Also, other modifiable risk factors promoting continued stress to the already fragile
vessel wall, or progressive endothelial damage such as in diabetes, should be treated
before re-initiating any kind of antithrombotic regimen after ICH.[64]
[65] Risk of ICH recurrence is higher in patients with lobar intracerebral bleeds compared
with those with hemorrhage in deep subcortical brain structures; nonetheless, resumption
of APT seems beneficial with regard to mortality and functional outcome even in patients
who suffered from a lobar bleed.[60] Superficial hemosiderosis and microbleeds are not visible on CT but their presence,
number, and location can be visualized on magnetic resonance imaging (MRI) and have
been linked to the risk of ICH,[61]
[66] A nearly fivefold increased ICH risk has been observed in patients with five or
more microbleeds.[67] However, even in patients with more than 20 microbleeds, the rate of ischemic stroke
exceeded that of ICH.[67] Whereas microbleeds in the white matter or basal ganglia result from small vessel
disease due to long-lasting arterial hypertension or diabetes, i.e., treatable conditions,
cortical microbleeds are found in patients with (untreatable) cerebral amyloid angiopathy
and are regarded as a precursor of atypical lobar bleeds. MRI might thus be superior
to CT when it comes to selection of ICH candidates for restarting antithrombotic therapy
([Fig. 3]).[68]
Fig. 3
Upper row: schematic drawing showing typical sequelae found in patients suffering from long-term
arterial hypertension, i.e., leukoaraiosis and deep microbleeds. (A) Leukoaraiosis on fluid-attenuated inversion recovery (FLAIR) MRI, (B) deep microbleeds within the basal ganglia and thalami, and (C) a hypertensive basal ganglia hemorrhage with intraventricular extension. Other typical
locations of hypertensive intracerebral hemorrhages are thalamic, pontine, and cerebellar
hemorrhages. Lower row: schematic drawing showing bleeding lesions indicative for cerebral amyloid angiopathy,
i.e., convexity subarachnoid hemorrhage (front left), lobar hemorrhage (front right), superficial hemosiderosis (back right), and cortical microbleeds (back left). (D) Superficial hemosiderosis, (E) cortical microbleeds, and (F) a hemorrhage within the left occipital lobe.
Strategies to Resume Antiplatelet Therapy
What Are the Patients' Overall Goals of Care?
The main goal of restarting APT is to prevent ischemic events without increasing bleeding
events and to reduce mortality and disability caused by recurrent stroke. For most
patients, APT, if indicated, should be restarted after most bleeding events, considering
that, for most of the therapies, fatal bleeding during APT is rare ([Table 1]) and so the risk–benefit assessment does not favor permanent discontinuation of
therapy. The mortality risk associated with major adverse events, e.g., MI or stroke,
outweighs the risk of dying from minor or moderate bleeding events (e.g., BARC 2 and
3a bleeding).[19] Interruption and resumption of APT should be individually tailored according to
bleeding severity/location, indication for antithrombotic therapy, risk for bleeding
recurrence, and thrombotic risk. APT should only be interrupted in case of major/severe
bleeding. In clinically relevant but not major bleeding events, therapy in patients
with clear indication for APT can be safely continued in most cases after short interruption.
There is no need for interruption of APT in patients with minor/trivial bleeding events.
Restarting/De-escalating Antiplatelet Therapy
The need for continuation of APT should be carefully assessed by weighing thrombotic
versus bleeding risk. The impact of early resumption (immediately after endoscopic
therapy) of low-dose ASA therapy was investigated in a small placebo-controlled trial
in 156 patients presenting with bleeding from peptic ulcer during ASA treatment. Within
30 days, the incidence of recurrent ulcer bleeding was 10.3% of patients who continued
ASA treatment versus 5.4% in patients receiving placebo (p = 0.25). All-cause mortality was markedly lower in the ASA group compared with the
placebo group (1.3 vs. 12.9%, p = 0.005). The low statistical power does not allow definite conclusions to be drawn
from this trial.[69] A recent meta-analysis including this RCT and observational studies reports inconsistent
evidence for a protective effect of aspirin resumption soon after nonvariceal upper
GIB.[70] Current guidelines recommend resumption of aspirin for secondary prevention within
3 to 7 days after upper GIB cessation.[71]
[72] The European Society of Gastrointestinal Endoscopy recently recommended determining
the time point of ASA re-initiation based on endoscopic classification of peptic ulcer
classification. In patients with low-risk endoscopic stigmata identified by Forrest
classification IIc or III, ASA should be restarted immediately. In patients with high-risk
stigmata (Forrest Ia, Ib, IIa, IIb), ASA should be restarted at day 3, given that
adequate hemostasis could be achieved.[73] Extrapolating data from patients undergoing polypectomy show a similar bleeding
risk between patients discontinuing and those continuing ASA.[74] The American College of Gastroenterology (ACG) recommends not to discontinue ASA
for secondary prophylaxis in patients with established cardiovascular disease in the
setting of lower GIB.[75]
In some situations, permanent cessation or de-escalation of DAPT may be considered.
In patients with major bleeding, the duration of DAPT can be shortened to 1 month
after elective PCI and to 6 months or less after ACS according to recent guidelines.[7]
[31]
[40] There are accumulating data from trials indicating safety of 1 month DAPT in high-bleeding-risk
patients treated with modern drug-eluting stents.[32]
[44]
[76]
[77]
[78] In patients with ACS undergoing PCI, the highest risk for recurrent cardiovascular
event lies within the first 3 months and cessation of DAPT within the first month
is associated with high risk of MACE and stent thrombosis.[9] Therefore, longer periods of discontinuation of DAPT (i.e., ≥5 days) should be avoided
in this vulnerable phase. Of note, the bleeding event itself can trigger a prothrombotic
condition by inflammatory processes, but also by the administration of blood products,
coagulation factors, and antidotes. Therefore, in patients with temporary enhanced
bleeding risk and high thrombotic risk requiring DAPT (e.g., patients with a recent
major bleeding and acute MI), antiplatelet compounds with short offset of action,
i.e., cangrelor might be a temporary option. However, there is only limited experience
with regard to safety and efficacy profile in bleeding situations. According to the
product labels, the use of cangrelor, eptifibatide, and tirofiban is off-label in
bleeding situations. Based on previous consensus documents, it is advisable to restart
DAPT within 3 days after moderate GIB.[9]
[73] DAPT should be interrupted for a longer period after major GIB and may be resumed
after 7 to 15 days in patients with low/moderate risk of recurrence.[73]
[79] There is no clear consensus if early re-initiation of a P2Y12 inhibitor and omission of ASA is associated with a favorable risk–benefit profile
early after major GIB. The ACG guideline recommends that, in patients on DAPT or SAPT
with clopidogrel, non-ASA-based APT should be resumed as soon as possible and at least
within 7 days based on multidisciplinary assessment of cardiovascular and gastrointestinal
risk. A second endoscopic examination may help to evaluate the further risk after
initiation of DAPT. DAPT should not be discontinued in patients with an ACS within
the past 90 days or coronary stenting within the past 30 days (strong recommendation,
low-quality evidence).[75] SAPT should be maintained if possible in most instances. Concomitant long-term therapy
with PPIs should be implemented if not already started.[73]
[80] In addition, de-escalation therapy to DAPT with clopidogrel in post-ACS patients
after a short period of DAPT with prasugrel or ticagrelor might be applicable to HBR
patients, which has shown to be beneficial regarding reduction of bleeding in recent
trials.[42]
[46]
[81] This includes de-escalation guided by platelet function or genetic testing, which
has demonstrated favorable effects in a recent meta-analysis[46] and could be particularly useful in patients with both high bleeding and ischemic
risk. Early switching to monotherapy after 1 to 3 months of DAPT to ticagrelor[41]
[45]
[82] or clopidogrel[44] monotherapy represents alternative strategies to reduce future bleeding events without
losing efficacy in selected patients. Availability of the specific antidote (i.e.,
antibody-based antidote bentracimab [PB2452] for ticagrelor) may be relevant when
choosing the oral P2Y12 inhibitor for long-term treatment in HBR patients. However, there are limited data
on efficacy and safety in acute bleeding situations and the antidote is not yet available.[83]
Restarting Combination Therapy (Anticoagulation + APT)
Triple therapy should be avoided in patients with atrial fibrillation (AFIB) undergoing
PCI who experienced a prior major bleeding event. The better safety of dual antithrombotic
therapy consisting of non-vitamin-K-antagonist oral anticoagulants (NOACs) plus a
P2Y12 receptor inhibitor compared with triple therapy[2]
[84] is established and highly recommended.[31]
[85] In HBR patients, switching from triple therapy (NOAC + DAPT) to dual therapy (NOAC + P2Y12 inhibitor) and from vitamin-K antagonists to NOAC are possible strategies to reduce
the bleeding risk. In case bleeding occurred under dual antithrombotic therapy (NOAC + SAPT),
dose reduction of NOAC to the lowest effective dose should be considered.[7] Although the preferred dosing for patients with AFIB undergoing PCI in general should
be the highest approved NOAC dose, if indicated, the current guidelines recommend
lowering the dose in HBR patients.[85] If risk for recurrent bleeding remains high even under a reduced NOAC dose, left
atrial appendage occlusion may be considered in patients to get along with APT only
for long-term treatment.
