Coagulation Disorders in the Neurocritical Patient

24 Coagulation Disorders in the Neurocritical Patient

Pablo Schoon ¹, Lilian Benito Mori ²

¹ Director of Intensive Care Unit. Hospital Interzonal General de Agudos “Prof. Dr. Luis Güemes”. Haedo, Buenos Aires State, Argentina

² Sub-Director of Intensive Care Unit, Hospital Interzonal General de Agudos “Prof. Dr. Luis Güemes”, Haedo, Buenos Aires State, Argentina

24.1 Introduction

Coagulation disorders in neurocritical or neurosurgical patients is a matter of significant importance. Hypercoagulability states in which intravascular thrombotic activity is increased, as well as deficits in hemostatic mechanisms leading to increased fibrinolysis and bleeding, compromise the evolution of these patients, overshadow the prognosis and worsen the results.

Clotting disorders may be the underlying cause of the neurocritical disease, as in spontaneous bleeding in patients with pre-existing coagulation disorders or in those under anticoagulation treatment. Also, an hypercoagulability state could be the etiology of an ischemic stroke. These disorders can increase secondary insult, promoting both the extent of intracranial bleeding or the development of intravascular thrombosis leading to brain ischemia. The frequency of the association between coagulation disturbances and neurocritical pathologies varies: the highest incidence (60%) is in patients with traumatic brain injury (TBI); 8-12% of cases of spontaneous intracranial hemorrhage (sICH) are related to anticoagulation treatment or coagulopathies; and <5% of patients with ischemic stroke are noted to have an hypercoagulability state.

24.2 Pathophysiology

In healthy individuals, a delicate balance is maintained between coagulation and fibrinolysis mechanisms to counteract excessive bleeding in the context of vascular lesion and inadequate thrombosis activity leading to vascular occlusion. These mechanisms, which include the activation of plasma and tissue factors, are closely regulated and activate only small proportions of clotting factors, since their concentration in a single millilitre of blood is sufficient to clot all the intravascular content within seconds.

In neurocritical patients, these mechanisms are often imbalanced, leading to inappropriate or excessive activation of the coagulation cascade, which results in complications of varying degrees of severity. In this chapter, we will describe the current state of knowledge of disturbances in coagulation-fibrinolysis mechanisms in neurocritical disease.

24.3 Traumatic Brain Injury

In TBI (traumatic brain injury) patients, clotting disorders are described as a combination of hypercoagulability and deficit in hemostasis, occurring simultaneously or successively.

It has been observed that, after the primary insult, tissue factors are released from platelets, white blood cells and endothelial tissue. This is the main trigger of the extrinsic mechanism of coagulation in which, through the triggering of the coagulation cascades, prothrombin is converted into thrombin, culminating in the formation of fibrin with the expected risk of thrombosis. While this “excessive” release of the tissue factor occurs in all trauma patients, the magnitude of this release is related more closely to brain injury than to shock, tissue hypoxia, or other extracranial injuries. It has also been reported that the degree of increase in tissue factor release is directly proportional to the extent of the brain injury. This trend to hypercoagulability triggers the activation of fibrinolysis mechanisms (increase in tissue factor inhibitor, protein C, protein S, antithrombin III, etc.). When fibrin production exceeds the capacity of these physiological mechanisms, an intravascular coagulation ensues in the small and medium-sized blood vessels. This intravascular coagulation can be generalized (disseminated intravascular coagulation leading to the development of extracranial organ failure), as well as localized. In necropsies of animals and patients, the latter has been observed in pericontusional and contusional areas where it leads to an increase in brain damage by extending tissue ischemia. Trauma patients may have a marked reduction in antithrombin III, protein C and S, resulting in thromboembolic or thrombotic complications rather than bleeding.

In TBI patients, coagulopathy may occur parallel with and/or secondary to the depletion of platelets and clotting factors due to hemorrhage, multiple transfusions and/or consumption by disseminated intravascular coagulation (DIC) in a setting of acidosis, hypothermia or hemodilution. Some studies have drawn attention to the occurrence of platelet dysfunction after TBI, more common than in trauma patients without TBI, and that it could have an impact on mortality. Whatever the mechanism involved, the presence of coagulopathy may extend the secondary injury by increasing bleeding in the intracranial lesions.

The incidence of coagulation disorders in TBI patients is related to the severity of the TBI, as categorized by the Glasgow Coma Score (GCS), and is correlated with evolution and prognosis. Various series have described an increase of up to 10 times in the risk of death in patients with coagulation disorders compared with those without such disorders and a greater risk of poor performance in the survivors with such disorders.

With regard to diagnosis, a good practice during the acute phase of TBI is daily assessment of activation partial thromboplastin time, platelet count and prothrombin time to precociously detect these disorders. Under clinical manifestations suggestive of DIC it is good practice to dose plasma concentrations of fibrinogen and fibrinogen degradation products. Specific determinations of clotting factors are not always available in all units, but evaluation is necessary in situations in which standard treatment does not control the coagulopathy state and a more precise assessment is needed.

24.3.1 Treatment

The therapeutic indication for a coagulation disorder in TBI patients should be aimed at controlling the predisposing factors that may have triggered it (hypothermia, acidosis or hypoxemia) and it also should be aimed at replacing the clotting factors. The administration of fresh frozen plasma (FFP) associated with cryoprecipitate solution is used for replacing clotting factors. The latter contains factor VIII and fibrinogen, both absent or at very low concentration in FFP, thus the desirability of their association. Platelets must be simultaneously reinstated, if necessary, according to the analytical.

24.4 Spontaneous Intracranial Hemorrhage

The main relationship between coagulation disorders and sICH is when it is secondary to anticoagulant therapy. This is the most serious and often fatal complication of such treatment. Much less frequently, the etiology of sICH is related to primary hematologic diseases: hereditary coagulation disorders and lymphoproliferative diseases.

The estimated incidence of sICH associated with the use of oral anticoagulants is 1.8% of patients treated with these drugs per year. This association raises the mortality of sICH by more than 10% compared to patients not under oral anticoagulant therapy. According to the 1999 SPIRIT study, those patients with a prior brain pathology and receiving oral anticoagulants have a higher risk of developing sICH. It has also been reported (Smith, 2002) that the presence of leucoaraiosis on brain images was an independent predictive factor for the development of sICH in patients treated with oral anticoagulants.

Gradient echo magnetic resonance imaging (MRI) studies have reported a high incidence of asymptomatic microhemorrhages in aged and/or hypertensive patients. It has been proposed that the use of oral anticoagulants could increase such bleeding events, making them symptomatic. This could be related to the finding that the majority of episodes of sICH, associated with the use of oral anticoagulants, occur within normal international normalized ratio (INR) therapeutic ranges, while higher values increase the risk.

The progression of bleeding of a sICH with expansion of the hematoma within 48 hours of the primary bleeding is seen in about 40% of patients with sICH. In cases associated with the use of oral anticoagulants or coagulopathies, this expansion could be more frequent, larger and extend beyond 48 hours, as has been observed in different published series.

Studies investigating the activation of hemostatic mechanisms in patients with intracranial bleeding have reported that such mechanisms are activated when there is subarachnoid or intraventricular blood. In these cases, there were significant increases in the thrombin-antithrombin and plasmin-antiplasmin complexes as compared to patients without subarachnoid blood and normal controls. According to these findings, there could be an increased risk, at least theoretically, of thrombotic events.

24.4.1 Treatment of Patients With sICH Associated With the Use of Oral Anticoagulants or Associated With Coagulopathy of Other Etiology

In addition to routine therapeutic measures in patients with sICH, in those cases in which it occurs in patients receiving oral anticoagulants or in those with clotting disorders, the treatment should include specific measures to reverse the anticoagulation state. These include the administration of vitamin K and clotting factors. This association is necessary, since the transfused factors have a median life of 5 to 60 hours, depending on each factor; therefore, they exert a rapid but transient effect. To ensure the reversal of anticoagulation over time, concomitant administration of vitamin K is required. Vitamin K begins to exert its effect on K dependent factor synthesis from 6 to more than 24 hours after intravenous infusion (doses of 5 to 20 mg). The intravenous administration of vitamin K solutions can potentially produce allergic reactions of varying magnitude. A slow infusion rate, not exceeding 1 mg per minute, is recommended. The risk of allergic and anaphylactic reactions is 3 in 10,000. Vitamin K given in either subcutaneous or intramuscular form will need longer to take effect.

There are two options in the administration of clotting factor solutions: fresh frozen plasma (FFP) or concentrate of prothrombinic complex (CPC). We will also consider the place of the cryoprecipitates (CP).

Since FFP has a low concentration of factors by volume, it must be administered in high volumes, which may be a limitation in some particular cases. One published study found that the infused volume needed to raise the INR 1.4 points ranged from 800 to 3500 ml. Of note is that the content and concentration of factors in such solutions are highly variable and depend also on storage time. Factor X has the greatest variability. This feature could normalize INR measurements, but it is associated with persistently low plasma levels of factor X and the risk of bleeding. Although this solution does not require compatibilization tests, it is advisable to infuse FFP obtained from donor blood of the same isotype as the recipient. The side effects associated with the use of these solutions include volume overload, allergic reactions, lung injury secondary to transfusion of blood products, and citrate toxicity. The recommend dose is 15-20 ml/kg body weight, but it may be necessary to increase this dose in cases of massive bleeding.

CPC contains K-dependent factors and proteins C, S and Z. It does not require compatibilization tests and, having a high concentration of clotting factors, it allows the infusion of smaller volumes than those needed for FFP.

Cryoprecipitates are solutions which contain factor VIII and fibrinogen, both factors scarce or absent in FFP. They are indicated in treating the specific deficits of these factors and also in cases of massive bleeding in anticoagulated patients, linking them to FFP.

As a recommendation and taking four clinical guidelines as reference, we suggest: vitamin K 5 to 10 mg IV + CPC 30-50 U/kg or FFP 15 ml/kg.

INR measurements are used for the follow-up, always taking into account prevention with respect to the X factor and the use of FFP alone.

Another more recent therapeutic option in the treatment of sICH is the use of recombinant activated VII factor (rVIIa), a potent trigger of the hemostasis cascade. It has been used in patients with hemophilia in intracranial or extracranial bleeding events. Experimental studies have shown that it has a rapid onset of action, mainly at the site of vascular injury. Also, it requires very low infusion volumes and it has good effectiveness and safety profiles. Because of these properties, it has been proposed and tested in patients with sICH to limit persistent bleeding and the growth of the hematoma, both features closely related to poorer outcomes. In a 2005 phase IIB study, rVIIa administration within 4 hours of the primary bleeding correlated with limiting the growth of the hematoma, lower mortality and improved outcome, assessed within 90 days after primary bleeding. But an increase in arterial embolic complications was also observed. According to this study, and the results of the FAST study, a Consensus Committee of experts concluded that the use of rVIIa in sICH limits the growth of the hematoma, but it has no beneficial effects on the survival rate at 90 days. So, rVIIa treatment remains confined to research studies and is not to be used in general practice.

There are few studies on the use of rVIIa in patients with sICH associated with oral anticoagulant therapy; mostly are retrospective with a small series of patients. It was warned that INR values might normalize after rVIIa administration, but, regardless of the persistently low values of other clotting factors, the risk of bleeding remains.

Some published clinical data, which might be confirmed by new studies, showed that reversal of the oral anticoagulant effect with the infusion of FFC or CPC could exacerbate perilesional edema by inducing the release of endothelial growth factor from “leakage” platelets, with an increase in capillary permeability. However, this phenomenon was not observed with the use of rVIIa in experimental studies.

24.5 Ischemic Stroke

A relationship between coagulation disorders and hemorrhagic complications in ischemic stroke is associated with the use of fibrinolytic therapy, but this issue is beyond the scope of this chapter.

Ischemic stroke might result from an hypercoagulability state. Several metabolic diseases or autoimmune disturbances may cause transient ischemic attack (TIA) or venous sinus thrombosis. Numbering among these disorders are: deficit of protein S, antithrombin III, protein C or plasminogen; prothrombin gene mutation; presence in the plasma of anticardiolipin antibodies or lupic anticoagulant factor; high plasma levels of homocysteine or lipoprotein a; and resistance of factor V to protein C. Specific laboratory studies for the detection of each of these disorders are needed when there is some degree of certainty that these mechanisms are involved in certain patients, according to the history and characteristics of the thrombotic event. Younger or pediatric patients with recurrent TIA events, or patients with recurrent venous thrombosis, or those with a family history of venous thrombosis, are likely to be considered affected by some kind of hypercoagulability disorder. Many of these disorders may be treated. A detailed discussion of these pathologies lies outside the scope of this chapter. An interesting and comprehensive review of these aspects was published by Rahemtullah and Van Cott in 2007.

24.6 Subarachnoid Hemorrhage

An increase in the activation of coagulation and fibrinolysis mechanisms has been described in patients with subarachnoid hemorrhage (SAH). It has been observed that the intensity of this activation (reflected in the increase in thrombin-antithrombin complexes, D dimer and fibrinogen concentrations) has a directly proportional relationship with the clinical severity of the bleeding, the amount of subarachnoid blood, the presence of intraparenquimatose hematoma or intraventricular blood. It is likely that these findings may explain what is observed in other studies which found a significant relationship between the magnitude in the elevation of these markers and a poor outcome. Therapeutic approaches to these disorders in patients with SAH have not been proposed.

24.7 Prophylaxis of Pulmonary Thromboembolism in Neurocritical Patients

Among the complications of critical care patients admitted to critical care units, thromboembolic disease and deep venous thrombosis (DVT) should be highlighted because of their high incidence and because they may result in severe consequences. The prophylactic therapy of DVT of the lower limbs is a routine issue in the basic care of all critical patients. Pharmacological therapy has shown greater efficiency for this purpose than mechanical methods such as elastic stockings, bandages or periodical/sequential compression pneumatic boots.

Critical patients with an acute neurological or neurosurgical disease constitute a special category. First, most of them have major risk factors for DVT such as prolonged immobility, coma, old age or associated injuries in the pelvis or lower limbs in trauma patients. On the other hand, pharmacological prophylactic therapy in these patients is reasonable thought as dangerous because it may impair intracranial bleeding, whether spontaneous or traumatic, and there is a risk of bleeding inside an ischemic stroke lesion.

We will summarize the current state of knowledge about the prevention of thromboembolic disease in neurocritical patients.

24.7.1 Ischemic Stroke and Spontaneous Intracranial Hemorrhage

The reported incidence of symptomatic DVT and/or pulmonary thromboembolism (PTE) in patients with ischemic stroke and sICH is 5-6%. However, the detection of asymptomatic DVT has been found to occur more frequently (from 5 to nearly 50%), depending on the method used for its diagnosis. In 2004, Kelly et al. published a study on 102 ischemic stroke patients who received prophylaxis with aspirin and compression stockings. They reported an incidence of DVT and symptomatic PTE of 3 and 5%, respectively, as diagnosed by MRI. But the incidence of DVT and PTE (symptomatic and more often asymptomatic) was 17.7 and 11.8%, respectively. At multivariate analysis, there was a significantly higher incidence in patients with a Barthel index ≤9 and an increased trend in patients over 70 years of age.

These findings, and those from other studies, underscore that we have to apply measures to prevent such serious complications. In recently published large series of patients, half of them who suffered a PTE event died as a consequence of it.

Prophylaxis

A Cochrane review, which included two randomized trials on the efficacy of compression stockings and pneumatic boots in the prevention of DVT in stroke patients, concluded that their use reduced the incidence of DVT, but the reduction was statistically non-significant. Furthermore, a randomized multicenter European study involving 2518 patients published in 2009 (The CLOTS trial collaboration) reported no significant reduction in the incidence of DVT with the use of elastic compression stockings, but it also reported a 4-fold higher incidence of cutaneous complications, including skin necrosis.

With regard to pharmacological prophylactic therapy, a systematic review assessing the efficacy of the use of antiplatelet drugs for the prevention of DVT/PTE concluded that their use reduces the risk of PTE but not the incidence of DVT.

Studies designed to assess the effectiveness of heparin administration in the acute phase of stroke to prevent stroke recurrence revealed that the incidence of PTE was reduced but with a concomitant significant increase in the risk of intra- and extracranial bleeding.

In 2007, Kamphuisen et al. published a systematic review of 17 studies involving 23,043 patients, in which low and high doses of low-molecular-weight heparin (LMWH) were compared with unfractionated heparin. They concluded that the use of low-dose LMWH was the therapeutic strategy that showed the best risk/benefit relationship in reducing DVT and PTE, without an increase in extracranial or intracranial hemorrhage complications.

In 2007 Sherman published the results of the PREVAIL study which included 1762 ischemic stroke patients unable to walk. Patients were randomized to receive, within 48 hours of the primary event, 40 mg of enoxaparin once daily versus 5000 IU of unfractionated heparin twice a day. The enoxaparin-treated group showed a significantly lower incidence of DVT and PTE, with no difference in the incidence of intracerebral bleeding between the two groups (1% in both groups). However, in those who received enoxaparin there was a significantly higher incidence of major extracranial bleeding (0.8 vs. 0%).

Sandercock, in a 2008 Cochrane review (9 studies involving a total of 3137 patients) compared different heparinoids versus unfractionated heparin administered within the first 14 days of an ischemic stroke event. LMWH was associated with a significant reduction in the incidence of DVT compared to standard heparin; however, Sandercock concluded that the data are insufficient to determine whether the use of this heparinoid reduces the risk of PTE or does not increase the risk of intra- or extracranial bleeding.

The 2007 American Heart Association and American Stroke Association (AHA-ASA) guidelines for the early management of adult stroke patients are based on published levels of evidence. The guidelines recommend: mobilize patients as early as possible; use subcutaneous LMWH in immobilized patients, but the ideal time for initiating therapy is unknown; aspirin is an alternative therapy but it is less effective than heparin; use mechanical compression methods when anticoagulants are contraindicated.

In 2008 Tetri et al. published a retrospective study involving 407 patients with sICH who survived 48 hours after the primary bleeding. They compared those who received 20 mg of enoxaparin once daily with those who did not receive prophylaxis during the acute stage. There were no differences in mortality at 3 months and there were no differences in the degree of growth of the hematoma between the two groups.

The update of the 2007 AHA-ASA guidelines for the management of adults with sICH based on the few articles published about this issue recommends the use of mechanical methods of intermittent compression. In those patients who remain immobilized or with motor deficit in the lower limbs within 3-4 days of the primary bleeding, low-dose LMWH may be used if there is no evidence of ongoing bleeding on the follow-up CT scans.

24.7.2 Traumatic Brain Injury

The incidence of asymptomatic DVT in patients with severe TBI is between 25 and 54%. The use of removable filters in the lower vena cava has been studied in a few trials. The analysis of these studies suggests that its use might be reserved only for those patients with known DVT and in which anticoagulation therapy is contraindicated.

With regard to mechanical or pharmacological prophylaxis, the published data on TBI patients are scarce.

A randomized study of patients with TBI and traumatic spinal cord injury (SCI) compared the use of intermittent pneumatic compression boots versus 40 mg enoxaparin once daily. The study found no significant differences in the incidence of DVT and PTE between the groups. Another retrospective study found that the use of LMWH, initiated within 72 hours of the primary injury, was not related to an increase in the size of intracranial hemorrhagic lesions.

In 2007 Cothren, published an observational study involving 6247 trauma patients (174 of which with TBI), who received 5000 IU dalteparine once a day. Dalteparine administration was begun when the patients showed hemodynamic stabilization and, in those patients with TBI, when the second CT scan at 12-24 hours after the primary injury showed no worsening in intracranial bleeding. The intracranial injury worsened in none of the patients and no death related to PTE or bleeding was recorded.

The 2007 Brain Trauma Foundation Guidelines based on published evidence recommend: use graduated compression stockings; the use of pharmacological therapy is known to potentially increase the risk of intracranial bleeding; there is not enough evidence to recommend any specific agent, doses or time after the primary injury to begin its administration.

24.7.3 Spinal Cord Injury

The incidence of DVT in SCI patients who not receive prophylaxis ranges between 7 and 100%, depending on injury severity, patient age, and the method to diagnosis DVT. Because of this high incidence, it is an accepted practice to use early prophylaxis in all of these patients. It was stated as a standard, with class I evidence, in the American Association of Neurological Surgeons Guidelines for the Management of Spinal Cord Injury. The guidelines also recommend extending prophylactic measures for 3 months after the primary injury, being reasonable to suspend them before such time if the patient has recovered motor function of the lower limbs.

With regard to which therapy is better, prolonged use of removable lower vena cava filters has been associated with some complications. A 2009 retrospective study suggests that the use of filters could increase the incidence of DVT. They are also indicated in patients under pharmacological prophylaxis who, despite this, underwent a thromboembolic event.

With regard to mechanical devices (graduated stockings, pneumatic boots), the results in some published studies suggest that they could be as effective as pharmacological measures. Others suggest that when associated with pharmacological therapy, their use can enhance its effectiveness. The above-mentioned guidelines recommend the combined use of mechanical devices and low-dose unfractionated heparin. We could add that, in those patients in whom anticoagulation therapy is contraindicated, it might be an indication for its use.

Finally, with regard to LMWH, several studies have found equal effectiveness by comparing the different products available. A 2009 systematic review including 30 studies concluded that LMWH is the best alternative because of its effectiveness and safety, whereas the use of unfractionated heparin has been associated with major bleeding complications.

24.7.4 Elective Neurosurgery

Published series of patients undergoing elective neurosurgical procedures (most of which affected by brain tumour) report an incidence of DVT between 19 and 54% and an incidence of PTE of around 5%. Despite these figures, there is reluctance to use pharmacological prophylaxis therapy in these patients because of the fear of an increased risk of bleeding. Two recently published surveys reported that only 32 to 33% of such patients receive pharmacological prophylaxis, although it is not definitively established whether therapy actually increases the risk of intracranial hemorrhage to any significant degree. Comparing study results is difficult because most studies are non-randomized and because spontaneous re-bleeding after elective neurosurgery has been reported to occur in more than 25% of cases. Also, the rate of reoperation because of post-operative re-bleeding ranges between 1 and 10%, even in patient series in which pharmacological prophylaxis for DVT was not administered.

In 2003, Gerlach published a study of a series of 2823 neurosurgical patients who received 0.3 ml of nadroparine in a single dose within 24 hours of surgery. The incidence of post-operative bleeding associated with clinical deterioration or which required a new operation was 1.5%. Also, Chibaro, in a 2008 study involving a series of 746 neurosurgical patients who received Tinzaparine, reported post-surgical intracranial bleeding in less than 1% of cases.

In 1998 Agnelli published a randomized study of 307 patients undergoing elective neurosurgery (90% involving the brain, 97% of which brain tumours) who received 40 mg enoxaparin once daily + graduated compression stockings versus graduated compression stockings + placebo. Enoxaparin was initiated within 24 hours of surgery and was continued for 1 week. The incidence of DVT and PTE was significantly lower in the enoxaparin group, and there was no increased incidence of major or minor bleeding between the groups.

Coincidentally with this last study, two meta-analyses (Iorio 2000 and Collen 2008), which included 7 and 30 studies, respectively, agreed that the use of LMWH proved effective in preventing DVT, without an excessive increase in the risk of bleeding. According to this evidence, the greatest benefits might be obtained by combining the use of mechanical compression devices with LMWH therapy, but the ideal time when these drugs should be initiated remains unknown.

General References

Only gold members can continue reading. Log In or Register to continue