Journal of Thrombosis and Thrombolysis

, Volume 30, Issue 3, pp 378–389

A clinical cardiology perspective of thrombophilias

Authors

    • Divisions of Cardiology and HematologyDuke University School of Medicine, Duke Clinical Research Institute
Article

DOI: 10.1007/s11239-010-0511-3

Cite this article as:
Becker, R.C. J Thromb Thrombolysis (2010) 30: 378. doi:10.1007/s11239-010-0511-3

Abstract

Thrombophilias, an inherited and/or acquired predisposition to vascular thrombosis beyond hemostatic needs are common in cardiovascular medicine and include systemic disorders such as coronary atherosclerosis, atrial fibrillation, exogenous obesity, metabolic syndrome, collagen vascular disease, human immunodeficiency virus, blood replacement therapy and several commonly used medications. A contemporary approach to patients with suspected thrombophilias, in addition to a very selective investigation for gain-of-function and loss-of-function gene mutations affecting thromboresistance, must consider prevalent diseases and management decisions encountered regularly by cardiologists in clinical practice. An appropriate recognition of common disease states as thrombophilias will also stimulate platforms for much needed scientific investigation.

Keywords

ThrombophiliaArterial thrombosisCardiovascular disease
Thrombophilias also referred to hypercoagulable states are the end result of diseases, disorders or conditions that heighten ones propensity to form blood clots within the venous, arterial and/or microcirculatory systems. Although classification schemes often consider inherited or acquired abnormalities of coagulation as a central theme for diagnostic evaluation and treatment (reviewed in references [1] and [2]) (Table 1), expanded paradigms are emerging which acknowledge common diseases, hospital-acquired contributors to include medications, environmental factors and evolutionary events within the cardiovascular system as both pre- and prothrombotic states.
Table 1

Relationship of risk factors and arterial thrombotic events

Arterial thrombotic risk factor

Associated with disease

Modifiable

Elevated homocysteine

Established

Plasma levels can be reduced with vitamin therapy, but clinical effect has not been demonstrated

Elevated CRP

Established

Plasma levels are modifiable with statin therapy with improved clinical outcomes

Presence of a LA

Likely, data limited

Not modifiable, but clear risk reduction with anticoagulation

Elevated levels of aCL

Possibly (elevated IgG)

Not modifiable, but clear risk reduction with anticoagulation

Elevated titers of β2-GPI antibodies

Possibly (β2-GPI antibody-dependent LA)

Not modifiable, but clear risk reduction with anticoagulation

Elevated fibrinogen

Established

Not modifiable, but clusters with other inflammatory markers and tobacco use

Elevated FVII

Not established

No

Elevated PAI-1

Not established

No

Factor V Leiden G1691A mutation

Possibly

Modest increase risk in combination with other cardiac risk factors, <55 years of age, and women

Prothrombin G20210A mutation

Possibly

Modest increase risk in combination with other cardiac risk factors, <55 years of age, and women

Protein C deficiency

Not established

No

Protein S deficiency

Not established

No

Antithrombin III deficiency

Not established

No

Platelet glycoprotein IIb/IIIa C1565T polymorphism

Not established

No

Platelet glycoprotein GPIb-IX-V C3550T

Not established

No

Platelet glycoprotein Ia/IIa C807T and G873A

Not established

No

Feinbloom and Bauer [2]

CRP C-Reacive Protein; LA Lupus Anticoagulant; aCL Anticardiolipin Antibody; GP1 Glycoprotein 1; PAI-1 Plasminogen Activator Inhibitor-1; G Glycoprotein

In broad terms, inflammation represents an overarching condition that is necessary, but not always sufficient to provoke thrombosis. More specifically, vascular injury, coupled with impaired healing and either delayed or maladaptive endothelial cell performance which precludes effective thromboresistance, is a requisite endophenotype governing clinical events.

It is very important to clarify that a clinical approach to thrombophilias should not be viewed as a call for redirecting scientific efforts designed to investigate complex genetic traits, genotype-phenotype relationships, vascular bed-specific events and gene expression profiles as platforms to understand mechanisms of disease and their capacity to provoke thrombosis (Fig. 1). Instead, retaining thrombophilias within the clinical arena serves as an important reminder that an explanation for thrombotic events, causing deep vein thrombosis, pulmonary embolism, ischemic stroke and/or an acute coronary syndrome may be closer than one often appreciates.
https://static-content.springer.com/image/art%3A10.1007%2Fs11239-010-0511-3/MediaObjects/11239_2010_511_Fig1_HTML.gif
Fig. 1

The biological response to many systemic disorders, traumatic events and environmental toxins is vascular injury and a resulting perception of a hemostatic challenge. In turn, a state of thrombotic preparedness ensures. An appropriate adaptive response achieves vascular repair and restoration of homeostasis—an appropriate balance between thrombotic potential and blood fluidity. The construct of genetic reserve may be an overarching theme governing adaptive, and in some instances, maladaptive response

Systemic diseases and conditions associated with thrombotic risk

Atrial fibrillation

Atrial fibrillation, an arrhythmia characterized by a cardiac-specific prothrombotic environment and resulting morbidity, mortality and staggering health care expenditures, is increasing at an alarming rate within the United States and other industrialized nations. Its prevalence reflects an aging population and epidemics of associated comorbid and prothrombotic illnesses, which include systemic hypertension [3], diabetes mellitus and atherosclerotic coronary artery disease.

The risk of cardioembolic stroke among patients with atrial fibrillation, on average 5% per year, but as high as 15 to 20% in selected high-risk subgroups, has placed antithrombotic therapy at the forefront of management. The heightened prothrombotic environment which characterizes atrial fibrillation (and atrial flutter) is underscored by numerous reports of recurring cardioembolic events despite anticoagulant therapy [4], the concomitant risk of thrombosis/thromboembolism occurring and originating, respectively, from within the small, low volume pressure right atrial appendage [5], and a distinctly poor outcome among patients having other conditions associated with thrombosis (e.g. myocardial infarction) who also experience atrial fibrillation [6].

Emerging evidence supports a pivotal role for inflammation in atrial fibrillation—both as a cause and an effect of the arrhythmia which contributes to recurrences as well as thromboembolic risk. Tissue samples obtained from the left atrial appendage of patients with atrial fibrillation and cardioembolic events reveal inflammatory cells, activated T lymphocytes, macrophages, and heightened expression of both tissue factor and von Willebrand factor [7]. Several markers of matrix metalloproteinase activity are decreased, an abnormality which could represent, at the tissue level, a predisposition to atrial fibrosis and thrombin generation [8]. In a study of 191 patients with non-rheumatic atrial fibrillation [9], interleukin (IL)-6 levels were elevated compared with age-matched controls in normal sinus rhythm. C-reactive protein levels are also increased [10], and may identify patients at increased risk for the subsequent development of atrial fibrillation [11] and failed attempts at cardioversion [12]. Atrial fibrillation following cardiac surgery correlates with postoperative leukocytosis, and polymorphisms of the -174G/C IL-6 promoter gene appear to modulate the inflammatory response in this setting, influencing the development of atrial fibrillation [13].

Emerging genetic and complementary molecular platforms are likely to play an increasingly important role in the understanding of atrial fibrillation and its association with cardioembolic and systemic thrombotic events. A genetic predisposition to atrial fibrillation is supported by the common occurrence of familial clustering, with up to 85% of individuals reporting one or more parents having experienced the arrhythmia [14]. Whether there is a concomitant accompanying predisposition to thrombosis in atrial fibrillation will require further investigation.

Vanhoutte et al. have encouraged a systems biology approach to atrial fibrillation [15]. Robust multichip average (RMA) normalized micro-array data from 10 patients with atrial fibrillation were analyzed using pathobiology-based probes for apoptosis, mitogen-activated protein(MAP) kinase (remodeling), OXPHOS (sustained myocardial work load) and glycolysis (ischemia). The analysis identified distinct patient groups for all probe sets, suggesting that molecular stratification and the identification of specific patient signatures, to include thrombosis phenotypes based on tissue expression [15] may be possible.

There are several important questions which future investigations, preferably undertaken in parallel with phase 2 and 3 clinical trials of antithrombotic therapy must address: (1) are there cultural/race-related differences in cardioembolic risk? [16]; (2) what is the contributing role (direct or indirect) of oxidative stress in thrombosis, inflammation and tissue remodeling? [17]; (3) what are the major obstacles to physicians prescribing anticoagulant therapy (when the risk of thromboembolism is known to be high)? [18]; and (4) what new direction in biomarker-related investigation, to include genetics, genomics, pharmacogenomics and gene expression profiles is required to achieve high yield prognostic and patient-specific information (which can be used for “real-time” management decisions)? [19].

Human immunodeficiency virus

According to the National Information Prevention Network (NPIN), an estimated 415,193 people in the United States alone are reported to be living with HIV and over 500,000 have died as a result of the disease [20]. Many people who are infected with HIV are unaware of their status and upward of 50,000 new infections occur yearly. The rate of HIV diagnosis for African Americans is 10-fold higher than among whites [20].

The hematologic complications of HIV infection are well known and include a heightened risk for venous, arterial (MI, stroke, peripheral vascular occlusion) and microvascular thrombosis (reviewed in reference [21]). In many, but not all cases the underlying propensity for thrombosis is attributable to autoimmune disorders and/or AIDS-related malignancies and infectious diseases [22]. Traditional coronary heart disease risk factors, coupled with HIV-specific lipid, metabolic and endothelial cell injury collectively create an environment which leads to high cardiovascular event rates [23]. Antiphospholipid antibodies contribute to the prothrombotic phenotype of HIV/AIDS [24, 25], as do acquired protein C and S deficiency [26], decreased antithrombin III levels, elevated factor VIII levels and advanced CDC-HIV classification [27]. A heightened platelet aggregation response to submaximal concentrations of epinephrine has also been reported [28].

HIV-related thrombophilia is not solely a condition of adults. Among children and adolescents, high viral load (HIV-RNA > 1,000 copies/ml) is associated with a reduction of protein S, protein C and antithrombin activities, and features of increased thrombin generation and fibrin formation [29].

Highly active antiretroviral therapy (HAART) has substantially increased life expectancy among individuals with HIV and is also associated with higher levels of protein C, free protein S and coagulant response to activated protein C, as well as decreased factor VIII activity, von Willebrand factor antigen and endogenous thrombin potential [30]. There is ongoing concern, however, that it also contributes to the growing number of vascular events. Mitochondrial toxicity of nucleoside reverse transcriptase inhibitors impairs glucose handling, causes insulin resistance and has been associated with atherothrombosis—in both the presence and absence of metabolic derangements (lipoatrophy/lipodystrophy) [3135]. Overall, the management of metabolic disorders and atherothrombosis among patients with HIV/AIDS is in a early stage of development [36] and must be investigated aggressively.

Autoimmune diseases

Autoimmune diseases, characterized by an immune-mediated abnormality of collagen, a highly prevalent protein of tendons, bones and connective tissue, has been associated with accelerated atherosclerosis, non-atheromatous vascular disease and thrombosis of the venous, arterial and microvascular circulatory systems. The autoimmune diseases encountered most frequently in clinical practice include rheumatoid arthritis, systemic lupus erythematosus, scleroderma (systemic sclerosis), dermatomyositis, Behcets disease and polyarteritis nodosa. The available information suggests that the risk of thrombosis among patients with autoimmune diseases is high, particularly in those with systemic lupus erythematosus [37] where there is a similar incidence of venous and arterial events.

Patients with rheumatoid arthritis, typified by high-grade systemic inflammation originating from joint synovia, are at substantial risk for coronary heart disease-related morbidity and mortality (reviewed in reference [38]). A pooled analysis of observational studies revealed a 70% increased risk of cardiovascular death which was minimally influenced by concomitant coronary heart disease risk factors [39]. Synovial tissue synthesizes and releases several inflammatory cytokines, including IL-1, IL-6 and TNF-α into the circulation. The importance synovial inflammation in both the pathogenesis and clinical expression of disease is supported by (1) a correlation between cardiovascular mortality and the number of inflamed joints; and (2) a reduction in cardiovascular mortality among patients treated with agents that reduce inflammation (e.g. methotrexate). Elevated fibrinogen, von Willebrand factor and D-dimer levels are also found among patients with rheumatoid arthritis. In addition, TNF-α, elevated substantially in this systemic disease, promotes tissue factor expression from monocytes [40], adding to the prothrombotic environment.

Systemic lupus erythematosus (SLE) is also characterized by inflammation within a variety of organ systems, including the cardiovascular system. Accelerated atherothrombosis is a relatively common cause of death (reviewed in references [41, 42]), with an alarmingly modest amount of information being available on preventing or attenuating the natural history of disease.

Patients with SLE are at risk for thrombotic events, particularly but not exclusively, in the presence of antiphospholipid antibodies, to include circulating lupus anticoagulants (reviewed in reference [42]). Recent data suggest that patients with a lupus anticoagulant have up to a 50% chance of experiencing a thrombotic event over the ensuing 20 years [43]. The presence of ox-LDL/β2-GPI complexes [44], also observed occasionally in patients with rheumatoid arthritis and systemic sclerosis, antibodies against β2-GPI and prothrombin (causing activated protein C resistance) [45] and circulating platelet-derived microparticles [46], also contribute to the prothrombotic phenotype of SLE. Smoking and disease activity are important determinants of thrombotic events [47], highlighting the importance of smoking cessation and aggressive disease management in these patients [48].

Not all patients with elevated antiphospholipid antibodies develop thrombotic complications. A study performed at Duke University Medical Center used gene expression profiles to accurately predict individuals at high risk for thrombosis [49]. An ability to discern thrombotic phenotypes using a genomic approach potentially represents a paradigm for effective management of disease among a wide array of “at risk” populations.

Patients with either limited or diffuse systemic sclerosis are at risk for venous, arterial and microvascular thrombotic events. The mechanisms are diverse and include vasospasm with secondary platelet-rich thrombosis, anti-endothelial cell antibodies and increased tissue-factor expression, fibro-proliferative disease, antiphospholipid antibodies to include a circulating lupus anticoagulant [5053] and thrombotic thrombocytopenic purpura (TTP) [54]. Medium-to-large vessel thrombosis, causing either an ischemic stroke or myocardial infarction is most often associated with antiphospholipid antibodies with or without a lupus anticoagulant [55]. Rarely, an associated cardiomyopathy with left ventricular mural thrombosis may underlie a sudden arterial occlusive event [56].

Activation of complement and increased levels of factor VIII with resulting thrombin generation, coupled with a small vessel vasculopathy likely contribute to microvascular thrombotic events among patients carrying a diagnosis of dermatomyositis [57, 58]. Medium-to-large vessel arterial thrombosis and venous thromboembolism are uncommon, and typically signal the presence of antiphospholipid antibodies and/or a circulating lupus anticoagulant.

Behcets disease, first described by a Turkish physician-dermatologist in 1937 is relatively common in the Middle East, Asia and Japan. The most frequent symptoms include oral and gential ulcers, uveitis, arthritis, skin rashes (erythema nodosum and pseudo-folliculitis), neuropathies, and vasculitis. The latter complication manifests as arterial narrowing, aneurysm formation, superficial and deep vein thrombosis, myocardial infarction, and stroke [59].

A similar pattern of vascular involvement is observed with polyarteritis nodosa; however, medium-sized vessel arteritis, to include the coronary and cerebrovascular beds, is more commonly observed than with dermatomyositis. In addition, polyarteritis nodosa has been associated with arterial aneurysm formation, representing a potential nidus for thromboembolism [60]. Antiphospholipid antibodies, as with other collagen-vascular diseases, substantially increase the risk of both venous and arterial thrombosis [61].

Inflammatory bowel disease

Patients with inflammatory bowel disease, including ulcerative colitis and Crohn’s disease, have an increased risk for thrombotic events of both the arterial and venous circulatory systems. Accelerated atherosclerosis of the aorta and coronary vasculature is not uncommon, and can be associated with MI and thromboembolic peripheral arterial occlusive events, respectively [62, 63].

The prothrombotic phenotype which accompanies inflammatory bowel disease is multifactorial and includes hyperhomocysteinemia secondary to impaired vitamin B12 and folate absorption [64], impaired fibrinolysis [65], platelet activation [66, 67] and concomitant hereditary thrombophilias [68].

Acute blood loss

The relationship between hemorrhagic complications of antithrombotic therapy among individuals with acute coronary syndromes remains a subject of clinical relevance. A prospective observational analysis from the CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress Adverse outcomes with early implementation of the American College of Cardiology/American Heart Association Guidelines) registry identified major bleeding as an independent predictor of mortality [69]. Similarly, the reduced number of deaths among patients who participated in the OASIS-5 trial (Organization to Assess Strategies in acute Ischemic Syndromes) [70] and received fondaparinux was almost entirely due to reduced bleeding. The relationship between major bleeding and mortality was observed at 9, 30 and 180 days, supporting an acquired and potent prothrombotic condition which may persist over time.

Independent risk factors for in-hospital and early (first 30 days) post-discharge bleeding include advanced age, female sex, renal insufficiency, anemia, elevated white blood count and use of one or more antithrombotic drugs [71].

The mechanism(s) underlying a consistently observed relationship between acute blood loss and ischemic-thrombotic events requires further investigation, but is likely multifactorial in origin [72], including maladaptive sympathetic responses, prothrombotic gene expression profiles, heightened inflammatory states [73] and possibly erythropoietin-mediated (endogenous or exogenous) endothelial cell activation and thrombin generation [74, 75].

Efforts to standardize the reporting of complications will undoubtedly facilitate an improved understanding of the mechanisms underlying adverse outcomes after bleeding [76].

Obstructive sleep apnea

A related area worthy of consideration in any discussion of cardiovascular thrombosis is obstructive sleep apnea and other disorders of breathing. Indeed, recent data support its direct association with cardiovascular death, MI, stroke and venous thromboembolism [77, 78]. Oxidative stress, increased plasma viscosity and impaired erythrocyte deformability may contribute to the prothrombotic phenotype [79].

Aging

A greater life expectancy has led to an increasing proportion of elderly persons in most developed and developing countries. Although age is a recognized risk factor for cardiovascular disease, the underlying mechanism(s) are poorly understood. Similarly, the relationship between aging and thrombotic events of the arterial (e.g. MI, ischemic stroke) and venous (e.g. deep vein thrombosis, pulmonary embolism) circulatory systems, a condition referred to as “prothrombotic imbalance or prothrombotic shift”, is largely undefined [80].

A meta-analysis of published studies correlating factor V Leiden, prothrombin G20210A and methylenetetrahydrofolate reductase mutations with arterial thrombotic events, failed to reveal a strong relationship among individuals greater than 55 years of age [81]. It is possible that arterial thrombosis, particularly within the coronary and/or cerebrovascular beds, is influenced modestly by small decreases in either the level or functionality of the vascular surface anticoagulant system. Similarly, the prothrombotic environment of aging may “dilute” the overall impact of some inherited thrombophilias [82, 83].

The relationship between age and thrombotic potential was investigated among 3,230 patients participating in the Framingham offspring study [84]. Increasing age correlated strongly with fibrinogen and von Willebrand factor levels, as well as measures associated with impaired fibrinolytic activity (plasminogen activator inhibitor-1).

Reduced fibrinolytic capacity has been observed among elderly men [85] suggesting that gene upregulation may represent a response to either inflammatory and/or hormonal mediators [86]. This unique environment may shed light on previously reported age-related changes in atheromatous plaque composition, characterized by decreased fibrous tissue and smooth muscle cell density, and increased macrophage density and MMP-9 content. The potential end-result of developing plaques with increased inflammatory and MMP activity is a greater propensity for disruption and localized thrombus formation, leading to clinical events [87].

One must also consider that endothelial cell dysfunction, as both a time-dependent and vascular toxin-determined phenomenon of aging, may not only represent a common underlying theme for atherothrombosis, but the effect (phenotype) of impaired vascular repair as well [88, 89]. Thus, obsolescence of endogenous progenitor cells may represent an important construct for understanding vascular events in aging [90], and their prevention using pharmacologic and/or cell-based therapies [91].

Obesity

Obesity is rapidly becoming a global epidemic of unparalleled proportion. While the initial alarm was sounded by the National Center for Health Statistics in 1994 (reviewed in reference [92]), changes in body mass index have accelerated rapidly in the past 5 years among children, adolescents and adults. In addition to its association with several atherogenic states, including type 2 diabetes mellitus, systemic hypertension and dylipidemia, obesity, in-and-of itself, is both proatherogenic and prothrombotic.

Obesity is an independent risk factor for deep vein thrombosis and pulmonary embolism [93, 94]. Data from the National Hospital Discharge Survey support a 2–3-fold increased risk for both obese men and women, particularly in those individuals less than 40 years of age [95]. Accordingly, thromboprophylaxis is an important consideration for practicing clinicians [96].

Beyond traditional risk factors for atherothrombosis observed frequently among obese individuals, one most consider additional acquired metabolic abnormalities and contributing disease states (reviewed in reference [97]).

Accumulating evidence indicates that obesity represents a state of low-grade inflammation which, in turn, leads to insulin resistance—both are associated strongly with systemic markers of inflammation [98] and atherosclerosis. Adipose tissue itself is composed of several cell types, including lipid-laden mature adipocytes, vascular stromal cells and macrophages. Co-culture of differentiated 3T3-L1 adipocytes and macrophages result in a marked upregulation of proinflammatory cytokines such as TNF-α and down-regulation of the anti-inflammatory and anti-atherothrombotic cytokine adiponectin [99]. Moreover, the inflammatory changes are augmented by adipose vascular stroma fraction, suggesting that inflammatory mediators participate in a maladaptive “paracrine loop” between adipocytes and macrophages.

Inflammation-sensitive plasma proteins (fibrinogen, orosomucoid, α1-antitrypsin, haptoglobin, ceruloplasmin) increase steadily with body mass index, and correlate with cardiovascular death, MI and stroke [100]. Soluble CD-40L, a marker of inflammation, platelet activation and prothrombotic potential, is elevated among obese patients, declining along with fasting insulin, MCP-1 and hs-CRP levels after bariatric surgery-associated weight loss [101]. Similarly, circulating procoagulant microparticles have been documented in obesity, with levels 3–4-fold higher than age-matched, non-obese controls [102].

Diabetes mellitus

Approximately 186,000 people under age 20 have diabetes and each year 15,000 people under age 20 are diagnosed with Type 1 diabetes. Among the adult population, current estimates suggest as many as 17,200,000 people may suffer from diabetes in the United States—nearly 8% of the population. Data from the Framingham Heart Study show a doubling of diabetes over the past 30 years and at least 65% of all individuals with diabetes die from cardiovascular disease, including MI or stroke [103]. Both arterial thrombosis and venous thromboembolism occur with increased frequency among individuals with diabetes, particularly those with poor glycemic control [104]. Several contributing effects have been proposed, including gene transcription of coagulation factors, facilitated assembly of coagulation factors on phospholipid surfaces and enhanced activation of glycosylated coagulation proteins.

Diabetes is associated with up-regulation of platelet-bound CD40L, enhanced P-selectin expression and increased soluble CD40L levels [105]. The inhibitory response of aspirin is attenuated in patients with diabetes [106] even among those with short periods of hyperglycemia [107]. The response to clopidogrel is also attenuated, with high on-treatment platelet reactivity in patients with coronary artery disease and concomitant renal insufficiency [108].

Last, acute insulin-induced hypoglycemia triggers activation and platelet monocyte aggregate formation [109].

Anemia: a procoagulant state?

Anemia, defined as a reduction in erythrocyte mass, is observed commonly among hospitalized patients, particularly those in whom chronic illness precedes their admission. The presence of anemia, whether accompanying medical illness, surgical procedures or trauma, portends a poor prognosis. Several common disorders characterized by acute or chronic states of inflammation and anemia include end-stage renal disease, malignancy, autoimmune disease, trauma, sepsis, and myocardial ischemia. Inflammatory mediators also exert prothrombotic effects on platelets, leukocytes, and the vascular endothelium, increasing the risk of both venous and arterial thrombotic events [110]. In chronic anemia, oxygen delivery is rarely compromised because of a gradual shift in the hemoglobinoxygen disassociation relationship and resulting preserved tissue perfusion. This may not be the case in more rapidly evolving conditions. Tissue hypoxia causes a prompt expression of inflammatory cytokines, which in turn, provoke erythropoietin synthesis and release. While increased erythropoiesis represents a physiologic response to anemia, hypoxemia, and tissue hypoperfusion, erythropoietin may exert direct prothrombotic effects, including platelet activation and induction of PAI-1 release—a procoagulant cytokine [111, 112]. The biological state of hemostatic preparedness, both at the genetic and protein levels, as an adaptive and teleological vital response to anemia is an area of active investigation.

Allogeneic blood product transfusion

The frequency of anemia among hospitalized patients has led to the liberal use of red blood cell transfusions. In 2004–2005, 3,500 patients received one or more red blood cell transfusions daily in US hospital intensive care unit (ICU) settings; totaling 1.25 million transfusions a year in the ICU alone. Rao and colleagues examined the potential impact of red blood cell transfusion in 24,111 patients with acute coronary syndrome (ACS). Several important observations from this post hoc analysis were made. First, there was a significantly higher 30-day all cause mortality and 30-day death or MI among patients who received transfusions. Second, increased risk persisted after adjustment for potentially confounding clinical variables (adjusted hazard ratio 3.94). Third, 30-day mortality was particularly high when transfusions were given to patients with hematocrits of 25% or above (compared to those with a hematocrit below 25%) [113]. The available data, derived predominantly from retrospective and modestly-sized prospective studies, raise suspicion for an association between allogenic blood transfusion and morbidity or mortality, yet the question remains whether the relationship is causal and how contributing factors intrinsic to stored blood may differ (or be manifest differently) based on coexisting illness and accompanying conditions, which themselves are prothrombotic [111120].

Pharmacologic factors

The potential for adverse events from drug therapy must be recognized by clinicians involved directly in patient care, regardless of whether they are considered common or extremely rare.

Intravenous immunoglobulin

Intravenous immunoglobulin (IVIg) preparations are used commonly in the treatment of patients with autoimmune, neurological and certain types of infectious diseases. Pooled from human plasma, IVIg is generally considered safe, with relatively minor adverse events reported in 10 to 20% of patients. Thromboembolic events, including VTE, MI and stroke can occur within hours to days of administration [121]. A review of 65 cases suggested the following : arterial thrombosis (MI, stroke) was four times more common than VTE and occurred in patients of advanced age and known atherosclerotic vascular disease; venous thrombosis was associated with concomitant factors contributing to venous stasis (obesity, immobility), and tended to be diagnosed later than arterial thrombosis. The overall mortality rate attributed to thrombosis was 20% [122]. Potential mechanisms include the introduction of clotting factors, vasoactive cytokines or factors that rapidly increase serum viscosity [123, 124].

Vascular endothelial growth factor antagonists

The contributing role of vascular endothelial growth factor (VEGF) to the growth and metastasis of highly vascularized tumors of the gastrointestinal tract has led to the development of a new class of chemotherapeutics. Although the benefit of anti-VEGF therapy is clear, early reports identified a 2-fold increase in serious arterial thromboembolic events among patients with colon carcinoma receiving Avastin® compared with standard chemotherapy [125]. The risk of VTE may also be increased [126].

While a cause-and-effect relationship has not been established, some have hypothesized that VEGF’s role as an antiapoptotic factor, and its ability to regulate endothelial cell proliferation and functionality, may precipitate plaque instability and impair vascular thromboresistance, respectively [127]. The presence of atherosclerotic disease, and the highly prothrombotic state of malignancy, provide a platform for the drug-related “triggering of events”.

Blood replacement products

The development of “bypassing agents”, for patients with hemophilia and circulating inhibitors to one or more coagulation proteins, such as activated prothrombin complex concentrate, factor Eight inhibitor bypassing activity, anti-inhibitor coagulant complex, vapor heated (FEIBA) and activated recombinant factor VIIa (rFVIIa), provide a fundamental frame work for extrapolating their use in the cardiovascular arena (reviewed in references [128, 129]).

FEIBA, used for over 30 years in the management of patients with hemophilia who have developed antibodies against f VIII or f IX, contains proenzymes of the prothrombin complex factors prothrombin, f VII, f IX and f X [130]. According to a pharmacovigilance database, thrombotic events occur in association with 4 of every 100,000 infusions [131] often when administered in doses exceeding current recommendations to patients with risk factors for cardiovascular disease.

Recombinant activated f VII has been shown to reduce mortality and improve functional recovery among patients with intracerebral hemorrhage treated within 3 h of diagnosis [132]. It may also be useful in the setting of warfarin-induced coagulopathy complicated by central nervous system bleeding [133] and either quantitative or qualitative platelet disorders [134, 135].

Off-label use of rVIIa is observed typically in settings characterized by bleeding risk (preventive strategy) or uncontrolled bleeding, with an adverse event rate of approximately 10%—a majority are the result of undesirable clotting [136] and are more likely in patients already at risk for thrombosis [137].

Antifibrinolytic therapy to reduce blood loss, transfusion requirement and re-operation following coronary arterial bypass grafting has received considerable attention over the past several years. Several small studies have documented reduced bleeding-associated events among patients receiving clopidogrel and aspirin who were treated with aprotinin [138]. A potential hazard associated with the routine use of aprotinin in high-risk patients (end-stage renal disease requiring dialyses, complex surgery, end-stage heart failure) has been brought to light [139, 140], with a concerning increase of MI, stroke and end-organ damage compared to alternative, and less expensive, antifibrinolytic agents (aminocaproic acid, tranexamic acid). Long-term follow-up subsequently identified an increased mortality among patients receiving aprotonin. The available data raise concern about the safety of aprotinin and warrant further investigation of dosing strategies (low dose versus high dose), causative mechanisms and the use of lysine analogues [141].

Cyclooxygenase inhibitors

The widespread use of anti-inflammatory agents that impact eicosanoid biosynthesis has led appropriately to increasingly in-depth investigation of their potential cardiovascular effects. Considering the available data, it is apparent that cyclooxygenase inhibition, both COX-1 and COX-2, creates a potentially harmful imbalance between thromboxane A2 and protacyclin on the vascular surface, augmenting prothrombotic potential [142]. The preexistence of endothelial dysfunction, as commonly accompanies atherosclerosis, shifts the balance further toward platelet adhesion and thrombosis, particularly with prolonged COX-2 antagonism ([143]; reviewed in reference [144]). The eventual trigger for arterial thrombotic events, including either MI or ischemic stroke, may be the direct effect of COX-2 inhibition on atheromatous plaques themselves through destabilization [145].

The prothrombotic effects attributable to COX-2 antagonists, perhaps most profoundly evident with increasingly selective agents, is highlighted by a report from the APPROVE off-drug extension [146], showing a 50% reduction in cardiovascular events following drug cessation.

The Multinational Etoricoxib and Diclofenac Arthritis Long-Term (MEDAL) Program included a pooled analysis of data from three trials of patients with either rheumatoid arthritis or osteoarthritis who were randomly assigned to etoricoxib, a COX-2 inhibitor, or diclofenac, a COX-1 inhibitor. A total of 34,701 patients were included in the analysis. Thrombotic cardiovascular events, occurring over an average of 18 months on treatment, were reported in 1.24 and 1.30 per 100 patient years, respectively [147].

The preferential use of non-steroidal anti-inflammatory drugs among patients at low risk for thrombotic events, combined with a strategy of selecting the least prothrombotic agent at the lowest effective dose for a brief treatment duration is advisable (reviewed in reference [148]).

Summary

Thrombophilias are common in cardiovascular disease. Although the search for inherited gain-of-function or loss-of-function genetic mutations should continue, an emerging paradigm will appropriately shift away from a focus on venous thrombosis-related genes to arterial thrombosis, genotype-phenotype relationships, reparative and thromboresistance regulatory protein gene anomalies and epigenetics as a means to understand the complex relationships between genes, gene expression, environmental factors, common endophenotypes and the clinical expression of disease. A clinicians approach to thrombophilias must remain focused on common diseases, hospital-acquired contributors and conditions which heighten thrombosis risk.

Copyright information

© Springer Science+Business Media, LLC 2010