Gastrointestinal Bleeding in Cardiac Patients
Gastrointestinal Bleeding in Cardiac Patients
In the 1980s and 1990s Helicobacter pylori and nonsteroidal anti-inflammatory drugs (NSAIDs) shaped our perspective of clinical GIB. GIB epidemiology was focused on peptic ulcer-related disease (PUD) and the impact of proton pump inhibitors (PPIs) and eradication of H. pylori in decreasing morbidity and mortality. Our attention then turned to the emergence of NSAID, aspirin (ASA) and anticoagulant-related lower GIB (LGIB), and the impact of age-related comorbidities, including ischemic heart disease and diverticulosis. The influence of comorbidity as an important independent risk factor for poor GIB outcomes, when compared with independent pharmacologic risk factors, has been recently explored and highlighted in the literature.
The most vulnerable population for upper (UGIB) and LGIB are elderly cardiac patients. These patients have multiple risk factors for GIB, including advanced age, a high burden of comorbidity and polypharmacy of anticoagulants, ASA and antiplatelet agents (i.e., antithrombotic agents). They often have preexistent mucosal defects that are at risk for bleeding, including diverticulosis, arteriovenous malformation and angiodysplastic lesions. It is estimated by 2030, 40.5% of adult Americans will have at least one cardiovascular disease requiring at least one antithrombotic agent. If projections for crude cardiovascular disease are superimposed on the United States population growth curve it is estimated that 27.2 million older Americans will have at least one cardiovascular condition treated by an antithrombotic medication for primary or secondary cardioprophylaxis by 2030.
The burden of cardiac comorbidities among the elderly often results in prescription antithrombotic agents in dual and triple combinations [i.e., complex antithrombotic therapy (CAT)]. A Spanish cohort study of 1219 patients following percutaneous coronary intervention demonstrated 96.9% were prescribed dual antiplatelet therapy [ASA + thienopyridine (e.g., clopidogrel)] at the time of discharge and only 76.6% were prescribed a PPI. Within 6 months, 2.2% had one or greater major GIB events, all of which were severe and required hospitalization. LGIB occurred more frequently than UGIB (P=0.012) and one case of LGIB was fatal. This study lacked the power to fully elucidate determinants of GIB. However, nonbleeding patients tended to be younger (66 versus 70 years; P = 0.06) and a prior peptic ulcer history and warfarin therapy were both associated with a three to four-fold increased risk of GIB.
The role of CAT in causing clinically significant GIB events has recently been described by Abraham et al. in Circulation. These data demonstrate the magnitude of risk and the frequency of events. The 1-year number needed to harm associated with ASA and an anticoagulant agent to incur one additional UGIB, LGIB, bleed-related transfusion or hospitalization is 93 [95% confidence interval (CI): 34–544], 18 (95% CI: 10–37), 51 (95% CI: 24–182) and 67 (95% CI: 30–214), respectively. Dual antiplatelet therapy with ASA and an antiplatelet therapy is associated with a number needed to harm of 93 (95% CI: 34–544) for UGIB, 18 (95% CI: 10–37) for LGIB, 51 (95% CI: 24–182) for bleed-related transfusion and 67 (95% CI: 30–214) for bleed-related hospitalization. Triple therapy with ASA plus anticoagulant plus antiplatelet is associated with a risk of harm in as few as 23 patients (LGIB), 52 (UGIB), 25 (bleed-related transfusion) and 45 (bleed-related hospitalization).
The clinical utilization of novel oral anticoagulants (NOAC; i.e., dabigatran, rivaroxaban, apixaban and edoxaban) has added a new dimension to the landscape of GIB. These agents differ from warfarin, a vitamin K antagonist, with regard to mechanism of action, metabolism, time to maximum effect, half-life, excretion and the ability to monitor the antithrombotic effect. Meta-analysis of randomized controlled data (RCT) demonstrate the risk of GIB with novel oral anticoagulants is 45% greater than warfarin alone, when used in patients with acute coronary syndrome (ACS), atrial fibrillation, deep venous thrombosis, pulmonary embolus and following orthopedic surgery [odds ratio (OR) 1.45; 95% CI: 1.07–1.97]. The bleeding risk for patients prescribed triple CAT using NOAC is also now emerging. NOAC prescribed in combination with ASA or a thienopyridine agent (i.e., clopidogrel, prasugrel or ticagrelor) after ACS, is associated with a three-fold increase in major bleeds (OR 3.03; 95% CI: 2.20–4.16), which translates to a number needed to harm of only 111 patients. In contrast, the same meta-analysis clarified the potential cardiac benefit as expressed by a number needed to treat of 77. These data demonstrate a very narrow threshold for risk-benefit that many elderly cardiac patients will have to navigate when treating multiple cardiac comorbidities in the future.
Gastrointestinal Bleeding Epidemiology: Old and New
In the 1980s and 1990s Helicobacter pylori and nonsteroidal anti-inflammatory drugs (NSAIDs) shaped our perspective of clinical GIB. GIB epidemiology was focused on peptic ulcer-related disease (PUD) and the impact of proton pump inhibitors (PPIs) and eradication of H. pylori in decreasing morbidity and mortality. Our attention then turned to the emergence of NSAID, aspirin (ASA) and anticoagulant-related lower GIB (LGIB), and the impact of age-related comorbidities, including ischemic heart disease and diverticulosis. The influence of comorbidity as an important independent risk factor for poor GIB outcomes, when compared with independent pharmacologic risk factors, has been recently explored and highlighted in the literature.
The most vulnerable population for upper (UGIB) and LGIB are elderly cardiac patients. These patients have multiple risk factors for GIB, including advanced age, a high burden of comorbidity and polypharmacy of anticoagulants, ASA and antiplatelet agents (i.e., antithrombotic agents). They often have preexistent mucosal defects that are at risk for bleeding, including diverticulosis, arteriovenous malformation and angiodysplastic lesions. It is estimated by 2030, 40.5% of adult Americans will have at least one cardiovascular disease requiring at least one antithrombotic agent. If projections for crude cardiovascular disease are superimposed on the United States population growth curve it is estimated that 27.2 million older Americans will have at least one cardiovascular condition treated by an antithrombotic medication for primary or secondary cardioprophylaxis by 2030.
The burden of cardiac comorbidities among the elderly often results in prescription antithrombotic agents in dual and triple combinations [i.e., complex antithrombotic therapy (CAT)]. A Spanish cohort study of 1219 patients following percutaneous coronary intervention demonstrated 96.9% were prescribed dual antiplatelet therapy [ASA + thienopyridine (e.g., clopidogrel)] at the time of discharge and only 76.6% were prescribed a PPI. Within 6 months, 2.2% had one or greater major GIB events, all of which were severe and required hospitalization. LGIB occurred more frequently than UGIB (P=0.012) and one case of LGIB was fatal. This study lacked the power to fully elucidate determinants of GIB. However, nonbleeding patients tended to be younger (66 versus 70 years; P = 0.06) and a prior peptic ulcer history and warfarin therapy were both associated with a three to four-fold increased risk of GIB.
The role of CAT in causing clinically significant GIB events has recently been described by Abraham et al. in Circulation. These data demonstrate the magnitude of risk and the frequency of events. The 1-year number needed to harm associated with ASA and an anticoagulant agent to incur one additional UGIB, LGIB, bleed-related transfusion or hospitalization is 93 [95% confidence interval (CI): 34–544], 18 (95% CI: 10–37), 51 (95% CI: 24–182) and 67 (95% CI: 30–214), respectively. Dual antiplatelet therapy with ASA and an antiplatelet therapy is associated with a number needed to harm of 93 (95% CI: 34–544) for UGIB, 18 (95% CI: 10–37) for LGIB, 51 (95% CI: 24–182) for bleed-related transfusion and 67 (95% CI: 30–214) for bleed-related hospitalization. Triple therapy with ASA plus anticoagulant plus antiplatelet is associated with a risk of harm in as few as 23 patients (LGIB), 52 (UGIB), 25 (bleed-related transfusion) and 45 (bleed-related hospitalization).
The clinical utilization of novel oral anticoagulants (NOAC; i.e., dabigatran, rivaroxaban, apixaban and edoxaban) has added a new dimension to the landscape of GIB. These agents differ from warfarin, a vitamin K antagonist, with regard to mechanism of action, metabolism, time to maximum effect, half-life, excretion and the ability to monitor the antithrombotic effect. Meta-analysis of randomized controlled data (RCT) demonstrate the risk of GIB with novel oral anticoagulants is 45% greater than warfarin alone, when used in patients with acute coronary syndrome (ACS), atrial fibrillation, deep venous thrombosis, pulmonary embolus and following orthopedic surgery [odds ratio (OR) 1.45; 95% CI: 1.07–1.97]. The bleeding risk for patients prescribed triple CAT using NOAC is also now emerging. NOAC prescribed in combination with ASA or a thienopyridine agent (i.e., clopidogrel, prasugrel or ticagrelor) after ACS, is associated with a three-fold increase in major bleeds (OR 3.03; 95% CI: 2.20–4.16), which translates to a number needed to harm of only 111 patients. In contrast, the same meta-analysis clarified the potential cardiac benefit as expressed by a number needed to treat of 77. These data demonstrate a very narrow threshold for risk-benefit that many elderly cardiac patients will have to navigate when treating multiple cardiac comorbidities in the future.