Haemostasis in critically ill patients with advanced chronic liver disease
Lead Research Organisation:
King's College Hospital
Department Name: Haematology
Abstract
What is the link between blood clotting and liver disease?
Liver disease is common and is now the largest cause of death in adults aged 30 to 59 years of age. Two serious complications of liver disease are bleeding and blood clots (thrombosis). Bleeding affects up to 1 in 10 adults with liver disease per year and blood clots up to 1 in 20 adults with liver disease per year. Most bleeding is due to pressure changes in the blood vessels in the abdomen leading to swelling of veins around the oesophagus (called varices).
What is known about blood clotting in liver disease?
The liver produces many of the substances involved in blood clotting, and patients with liver disease often have marked abnormalities in blood clotting measured with routinely available tests. This is because routinely available tests only measure the effect of proteins which increase blood clotting (pro-coagulants). In liver disease, the proteins which reduce blood clotting (anti-coagulants) are also reduced and overall, clotting is thought to be 'rebalanced'. Specialist tests (thrombin generation and thromboelastography) which are able to measure either the effect of the body's anti-coagulants or blood cells demonstrate that clotting is actually normal or increased in most patients with liver disease.
Why are blood tests to measure blood clotting in liver disease important?
Current routine tests are poor at predicting bleeding and clotting complications in patients with liver disease. Because the blood clotting tests in patients with liver disease look abnormal, treatments like transfusion of plasma (blood clotting proteins), platelets (small cells important for blood clotting) or fibrinogen (the final protein needed for blood clot formation) may be given to try and improve the blood test results, even though this might not reduce the risk of bleeding. In patients who are bleeding, the treatment approach is based on studies in different patient groups such as trauma where plasma transfusion has been shown to improve outcomes. In liver disease, studies show treatment with plasma can worsen bleeding from varices by increasing the pressure in the veins. Patients with liver disease are also at increased risk of thrombosis; most patients in hospital are given medications to reduce the risk of thrombosis ('blood thinners' or anticoagulants) but due to the changes seen on routine laboratory tests, doctors sometimes worry these medications might increase the risk of bleeding. Balancing the risks of bleeding and thrombosis and deciding which blood test results should be treated can be very difficult. This is particularly important in seriously ill patients as they are at higher risk of bleeding and thrombosis, and often need multiple procedures.
What will I do?
I will look at specialist blood tests (thrombin generation, thromboelastography, neutrophil extracellular traps) in patients with liver disease admitted to intensive care to see whether these tests could be used routinely to guide the need for transfusion and predict the risk of bleeding or thrombosis. I will also look at whether these tests remain the same over time and how transfusion changes the results.
Why is this research important?
As current tests are not helpful, many patients get transfusions when they probably don't need to. This is wasteful, costly and puts patients at risk of side effects. This is important both for patients with liver disease and bleeding (4000/year) and those having procedures/operations. Better tests will reassure doctors when transfusion is not needed and will improve management of bleeding complications and prevention of thrombosis in hospital.
Liver disease is common and is now the largest cause of death in adults aged 30 to 59 years of age. Two serious complications of liver disease are bleeding and blood clots (thrombosis). Bleeding affects up to 1 in 10 adults with liver disease per year and blood clots up to 1 in 20 adults with liver disease per year. Most bleeding is due to pressure changes in the blood vessels in the abdomen leading to swelling of veins around the oesophagus (called varices).
What is known about blood clotting in liver disease?
The liver produces many of the substances involved in blood clotting, and patients with liver disease often have marked abnormalities in blood clotting measured with routinely available tests. This is because routinely available tests only measure the effect of proteins which increase blood clotting (pro-coagulants). In liver disease, the proteins which reduce blood clotting (anti-coagulants) are also reduced and overall, clotting is thought to be 'rebalanced'. Specialist tests (thrombin generation and thromboelastography) which are able to measure either the effect of the body's anti-coagulants or blood cells demonstrate that clotting is actually normal or increased in most patients with liver disease.
Why are blood tests to measure blood clotting in liver disease important?
Current routine tests are poor at predicting bleeding and clotting complications in patients with liver disease. Because the blood clotting tests in patients with liver disease look abnormal, treatments like transfusion of plasma (blood clotting proteins), platelets (small cells important for blood clotting) or fibrinogen (the final protein needed for blood clot formation) may be given to try and improve the blood test results, even though this might not reduce the risk of bleeding. In patients who are bleeding, the treatment approach is based on studies in different patient groups such as trauma where plasma transfusion has been shown to improve outcomes. In liver disease, studies show treatment with plasma can worsen bleeding from varices by increasing the pressure in the veins. Patients with liver disease are also at increased risk of thrombosis; most patients in hospital are given medications to reduce the risk of thrombosis ('blood thinners' or anticoagulants) but due to the changes seen on routine laboratory tests, doctors sometimes worry these medications might increase the risk of bleeding. Balancing the risks of bleeding and thrombosis and deciding which blood test results should be treated can be very difficult. This is particularly important in seriously ill patients as they are at higher risk of bleeding and thrombosis, and often need multiple procedures.
What will I do?
I will look at specialist blood tests (thrombin generation, thromboelastography, neutrophil extracellular traps) in patients with liver disease admitted to intensive care to see whether these tests could be used routinely to guide the need for transfusion and predict the risk of bleeding or thrombosis. I will also look at whether these tests remain the same over time and how transfusion changes the results.
Why is this research important?
As current tests are not helpful, many patients get transfusions when they probably don't need to. This is wasteful, costly and puts patients at risk of side effects. This is important both for patients with liver disease and bleeding (4000/year) and those having procedures/operations. Better tests will reassure doctors when transfusion is not needed and will improve management of bleeding complications and prevention of thrombosis in hospital.
Technical Summary
Better tests to evaluate haemostasis in patients with chronic liver disease (CLD) are required to reassure clinicians when blood component support is not required and to guide transfusion in the event of bleeding. I am acutely aware of this both as a clinician specialising in thrombosis and haemostasis at one of the largest regional liver centres in Europe, and also as a working group member for the ISTH on 'Haemostasis in patients with liver disease'.
Aim: to identify novel methods to assess haemostasis in critically ill patients with CLD, and determine their association with bleeding and thrombosis.
Objective 1:To evaluate the clinical utility of thrombin generation (TG) and thromboelastography (TEG) in critically ill patients with CLD in reducing blood component use
Objective 2: To identify clinical and laboratory risk factors (including TG, TEG, clot lysis time [CLT] and neutrophil extracellular traps [NETs]) for bleeding and thrombosis
I will set up a prospective observational multicentre cohort study to recruit 250 patients with CLD admitted to critical care, and 30 healthy controls. Blood samples and clinical data will be collected on admission, day 1 and day 5 with clinical outcomes recorded up to day 28 (or liver transplant, discharge or death, whichever is sooner). Major bleeding will be defined by the HEME tool and thrombosis will require confirmation with objective imaging. Thrombin generation will be measured in both plasma and whole blood, along with TEG, CLT and NETs. We will establish the proportion of patients with normal haemostatic profiles on admission, the stability of profile over the initial 5 days of admission and the effect of blood components transfused on haemostatic profile. Associations between clinical risk factors, admission TG and TEG parameters with later bleeding/thrombosis will be evaluated, as will the relationship between markers of NETs and aetiology of CLD, presence of sepsis on admission and later thrombosis.
Aim: to identify novel methods to assess haemostasis in critically ill patients with CLD, and determine their association with bleeding and thrombosis.
Objective 1:To evaluate the clinical utility of thrombin generation (TG) and thromboelastography (TEG) in critically ill patients with CLD in reducing blood component use
Objective 2: To identify clinical and laboratory risk factors (including TG, TEG, clot lysis time [CLT] and neutrophil extracellular traps [NETs]) for bleeding and thrombosis
I will set up a prospective observational multicentre cohort study to recruit 250 patients with CLD admitted to critical care, and 30 healthy controls. Blood samples and clinical data will be collected on admission, day 1 and day 5 with clinical outcomes recorded up to day 28 (or liver transplant, discharge or death, whichever is sooner). Major bleeding will be defined by the HEME tool and thrombosis will require confirmation with objective imaging. Thrombin generation will be measured in both plasma and whole blood, along with TEG, CLT and NETs. We will establish the proportion of patients with normal haemostatic profiles on admission, the stability of profile over the initial 5 days of admission and the effect of blood components transfused on haemostatic profile. Associations between clinical risk factors, admission TG and TEG parameters with later bleeding/thrombosis will be evaluated, as will the relationship between markers of NETs and aetiology of CLD, presence of sepsis on admission and later thrombosis.
