Dr. Artur Fedorowski earned his MD and PhD in internal medicine and cardiology from the Wroclaw Medical University in Poland. Since 2013, he has worked as an Associate Professor at Lund University, and since 2016, he has been a Fellow of the European Society of Cardiology. Dr. Fedorowski has vast experience in cardiology and cardiovascular medicine. In this video Dr. Fedorowski speaks about Thrombosis and fibrinolysis in atherosclerotic cardiovascular disease.
Link to Article:
Walter Bradford Cannon (1871–1945) developed the term "homeostasis," which refers to "any self-regulating process by which biological systems strive to preserve stability while responding to conditions that are optimal for survival." If homeostasis is achieved, life continues; if it is not, disaster or death follows' (Encyclopaedia Britannica; www.britannica.com). Haemostatic capacity, which prevents bleeding while preserving unobstructed blood flow, is one of the most important aspects of human homeostasis. Platelets and the coagulation system are two key components of haemostasis, working together to form a haemostatic plug when vascular continuity is interrupted. Coagulation and platelet activity are turned off by circulating inhibitors and proteolytic feedback loops to reduce the danger of arterial occlusion. 2 Otherwise, uncontrolled haemostatic activity will cause thrombo-embolic problems, leading to 'disaster or death' in the worst-case scenario. Prothrombotic and fibrinolytic forces are thus intertwined, resulting in a homeostatic steady state that responds to vascular emergencies.
DeFilippis et al.3 present a prospective study based on a multi-ethnic cohort of 5800 adults (mean age 63 years) without a history of atherosclerotic cardiovascular disease (ASCVD) from six communities in the United States in this issue of the European Heart Journal. The participants were recruited between 2000 and 2002 and monitored for 15 years, during which time 15% of them had a cardiovascular event. Blood samples were taken at the start of the study to look for circulating prothrombotic and fibrinolytic factors to see if there was a biomarker pattern that could predict incident ASCVD. The authors' technique is unique in that it uses two composite variables called thrombotic and fibrinolytic factors. Five biomarkers contributed to the thrombotic factor [fibrinogen, plasmin–antiplasmin complex, factor VIII, D-dimer, and lipoprotein(a)], and two contributed to the fibrinolytic factor [plasminogen and oxidized phospholipids on plasminogen (OxPL-PLG)]. The authors discovered that a higher thrombotic factor score and a lower fibrinolytic factor score suggested the highest risk of incident atherothrombotic events (Graphical Abstract).
Higher levels of fibrinogen, D-dimer, plasmin–antiplasmin, factor VIII, and Lp(a) have previously been linked to an increased risk of CV events, therefore the findings of this significant study3 were not entirely unexpected.
4–7 Importantly, circulating levels of both plasminogen and OxPLs, which indicate reduced fibrinolytic activity, were significantly lower in patients with incident myocardial infarction at baseline compared to stable coronary artery disease patients and unaffected subjects, but increased immediately after percutaneous coronary intervention. 8 What's new is that assessing the balance between the competing processes of thrombosis and fibrinolysis should help forecast CV events more precisely. The authors underline that when the two biomarker sets are evaluated combined, they provide additional information. It was proved in a statistical model where the relative risks were increased once both factors were input at the same time. Notably, there was a moderately positive association between these two variables, implying that increased thrombotic propensity is frequently counterbalanced by increased fibrinolytic propensity. When thrombotic propensity and fibrinolytic propensity were uneven, i.e. when an elevated thrombotic factor was associated with a low fibrinolytic factor, the risk of CV events was highest. As a result, patients with high fibrinolytic factor and low thrombotic factor values had the lowest risk of CV events. It's worth noting that the two factors had no effect on coronary artery calcification.
Adding thrombotic and fibrinolytic components to a multivariable Cox model using variables from the American Heart Association (AHA)–American College of Cardiology (ACC) CV disease risk prediction calculators resulted in a minor (albeit statistically significant) improvement in model fit (C-index increase from 0.718 to 0.721). As a result, the proposed model based on integrated coagulation biomarkers has the potential to provide extra information on CV risk prediction beyond what can be determined only from traditional risk variables. However, the question arises as to whether measuring the balance between thrombosis and fibrinolysis has therapeutic value when applied to well-established clinical risk models in a broad and perhaps arbitrary manner. Indeed, a recent study of community-dwelling middle-aged people found that a high CHA2DS2-VASc score, which was designed for patients with atrial fibrillation, was also a sensitive measure for predicting bad CV outcomes in people who did not have atrial fibrillation. 9
DeFilippis et al.3 claim that the new thrombotic and fibrinolytic factors allow for a more complete assessment of the balance between thrombosis and fibrinolysis, which determines the risk of atherothrombosis, and that the new factors have promise for predicting an individual's propensity to form thrombus, based on their findings. As a result, the model should integrate into precision medicine and has the ability to select individuals for antithrombotic therapy by weighing the risk of iatrogenic bleeding against the advantages of reducing atherothrombotic events. For primary CV prevention, such a tailored medicine strategy appears to be quite appealing. Despite this, the path from epidemiological observation to clinical application is typically protracted, with just a few biomarkers making it into current clinical practice. The primary difficulty here will be to show that the suggested composite biomarker prediction model is capable of selecting candidates for more or less aggressive CV disease prevention, beyond what can be clinically determined in conjunction with traditional risk assessment scores.
However, it should be highlighted that other major atherothrombotic risk factors, such as platelet indices and platelet reactivity, as well as atherosclerotic indicators, were not considered in DeFilippis et al.3's investigation (Graphical Abstract). Individuals with an increased platelet count at baseline had a significantly greater risk of death and severe CV events throughout a 16-year follow-up in the prospective Malmö Diet and Cancer cohort, regardless of the presence of CV disease. 10 Higher platelet aggregability, as well as pathways connected to atherosclerosis onset, progression, and complications in people with high platelet counts, could explain these findings. Furthermore, a study of the same cohort found that clustering of blood cell count abnormalities (e.g., anemia, leucocytosis, and thrombocytosis), a condition commonly seen in people with a high inflammatory milieu, poor health, or frailty, was linked to increased mortality and the risk of future CV events. 11 These findings, together with the findings of DeFilippis et al.3, raise the question of whether anticoagulants such as low-dose rivaroxaban12, rather than aspirin13, should be explored in primary CV prevention in those with a high prothrombotic profile.
DeFilippis et al.3's study has several clear strengths: robustness of data gathered prospectively from a large, representative population; independent and meticulous event adjudication; long follow-up duration; and long-term stability of healthcare and reporting systems. However, there are some significant restrictions. First, because the investigators only looked at baseline blood samples, the dynamic nature of atherothrombotic biomarkers was not investigated. Second, data on platelet reactivity, which is important for assessing atherothrombotic risk, were not available, and aetiological subtyping of CV events (for example, atherothrombotic vs. non-atherothrombotic myocardial infarction) was not done. The danger of residual confounding and a lack of detailed information on concomitant disorders or undiscovered diseases that could cause changes in coagulation patterns at baseline are two other drawbacks. Importantly, the relationships found by DeFilippis et al.3 do not prove causation and must be confirmed in separate cohorts. Furthermore, due to the demand for seven laboratory measurements, the model's relevance in the real world should be acknowledged. Finally, it will be interesting to see how thrombotic and fibrinolytic variables vary over time and how these changes relate to cardiovascular events.
Taken together, our findings support the notion that the use of multi-biomarker technologies can help to improve CV risk prevention even more. Different mechanisms (e.g. thrombosis, inflammation, and lipid plaque development) may coexist in the pathophysiology of atherothrombosis, with varied roles for each of these mechanisms in different individuals (Graphical Abstract). As a result, the best prediction model would combine data from the coagulation system, lipid metabolism, and inflammation. As a result, a two-step approach may be beneficial: first, individual clinical risk stratification to identify people at moderate to high risk of CV events, then multi-biomarker examination to determine which pathogenetic process (thrombosis vs. inflammation vs. lipid problems) is dominant. The latter may help the practitioner choose whether preventive strategies should be used on an individual basis. A machine learning approach may also be effective for developing multi-parametric models due to its accuracy in analyzing competitive hazards. 14 Another major concern is whether prothrombotic vs. fibrinolytic activity should be assessed separately.