Which traumatic brain injury patients should be treated with anticoagulants and when?

Abstract

Anticoagulation to prevent venous thromboembotic events in head trauma is a controversial intervention. Neurosurgeons have generally been opposed to the idea. We present our view of the safety and efficacy of chemoprophylaxis for patient with traumatic brain injury, as well as our recommendation for such intervention.

Keywords::

• anticoagulation
• heparin
• Lovenox
• TBI
• trauma

The question of anticoagulation use in patients with traumatic brain injury (TBI) has been debated by trauma surgeons and neurosurgeons for decades. Trauma surgeons feel that anticoagulation is relatively safe and effective in preventing thromboembolic events (VTE), including deep venous thrombosis (DVT) and pulmonary embolism (PE), based on extensive experience in patients with injuries not involving the brain. Many neurosurgeons approach the concept of chemoprophylaxis for VTE for patients with TBI with trepidation, based on fear that anticoagulation medications will either cause a new hemorrhagic process to occur or worsen intracranial hemorrhage (ICH) seen on initial presentation. The most recent guidelines for the management of severe TBI (2007 edition) address the role of chemoprophylaxis as an option for treatment, but warn of increased risk of expansion of ICH with their use [1]. However, recent studies support the use of chemoprophylactic anticoagulant medications in patients with TBI, regardless of the severity of the injury, as a safe and effective treatment option.

TBI can cause either a prothrombotic or a coagulopathic state. Secondary injury can occur due to the production of emboli or the progression of hemorrhagic brain lesions [2]. A number of proposed mechanisms include the release of tissue factor, hyperfibrinolysis, hypoperfusion triggering the protein C pathway, disseminated intravascular coagulation and platelet dysfunction [3]. Significant systemic morbidity and mortality may occur with either prothrombotic or coagulopathic states including DVT, PE, myocardial infarction, ischemic stroke and systemic hemorrhage. Nevertheless, based upon a review of available literature and our own research [4], we favor chemoprophylactic anticoagulation in the treatment of TBI patients who have a stable 24-h post-injury head computed tomography (CT).

The incidence of DVT in TBI patients with no mechanical prophylaxis or anticoagulation is relatively high. A recent prospective observational study of 36 TBI patients showed a DVT rate of 6% and PE rate of 6% [5]. In another study, a subset of 57 patients with TBI had a 16% incidence of DVT in patients with no mechanical prophylaxis, an 8.8% incidence of DVT in patients treated with compression stockings and a 9% incidence with compression stocking and intermittent pneumatic compression [6]. Given the extremely low risk of mechanical prophylaxis and that VTE are relatively common, mechanical DVT prophylaxis in the setting of TBI is highly appropriate. However, as the above data indicate, mechanical prophylaxis by itself is insufficient to combat the morbidity and mortality associated with VTE.

Two questions need to be answered to create comfort for neurosurgeons with the decision to provide anticoagulation chemoprophylaxis to patients with TBI. First, is anticoagulation safe? Second, is anticoagulation effective?

Safety first! Kleindienst et al. [7] utilized low molecular weight heparin (certoparin 3000 units anti-factor Xa subcutaneously [sc.]) 24 h after stable head CT in 344 TBI patients. Nine patients (2.6%) had expanding intracranial hematoma, eight of whom required further surgical intervention. However, four of these patients received anticoagulation prior to a stable follow-up head CT. Minshall et al. [8] cited 5 and 12% rate of ICH progression in TBI patients evaluated with follow-up head CT, if treated with low molecular weight heparin (enoxaparin 30 mg sc. twice per day) versus unfractionated heparin 5000 units sc. three-times per day, respectively. This study also suggested potential safety concerns for chemoprophylaxis. However, the follow-up head CT scans that documented progression were only obtained when clinically indicated. The true incidence of hemorrhagic progression would likely have been much lower if all patients had follow-up imaging after initiation of chemoprophylaxis.

A number of other studies demonstrate the safety of chemoprophylaxis. A retrospective review of 812 patients admitted to a Level I trauma center between 2006 and 2008 found that the 3% incidence of hemorrhagic progression in those who received chemoprophylaxis was significantly different than the 6% rate in those who did not. Interestingly, the VTE rate in this study was only 1.7% [9] as compared with 6–9% rate observed by Turpie et al. [6]. Furthermore, those patients with isolated head injury were less likely to receive chemoprophylaxis than others (13 vs 68%) [9], perhaps indicative of admission to the neurosurgical service and trepidations of the neurosurgeons for the isolated head injury group, one of the reasons for presenting our view of chemoprophylaxis in this setting. A 2008 study by Norwood et al. [10] corroborates these data with their cohort of 525 TBI patients initially treated with enoxaparin 30 mg sc. twice per day 24 h after arrival in the ER, in which 3.4% had ICH progression, with only 1.1% requiring a change in management. Chemoprophylaxis was thought to contribute to only one death (0.2%). At University of Missouri Hospital and Clinics, we prospectively evaluated 129 patients treated with a protocol of chemoprophylaxis for VTE after a stable 24 h head CT and compared them with a previous cohort of 107 patients not treated with the protocol. ICH progression in the cohort with routine chemoprophylaxis was 0.78%, which was essentially no different than in the cohort without chemoprophylaxis (2.8%). Of note, in our study, all patients had head Abbreviated Injury score >3, all patients had a follow-up head CT 48–72 h after initiation of chemoprophylaxis and chemoprophylaxis was intended to be initiated after a stable head CT 24 h after injury [4]. These results are consistent with those of Cothren et al. [11], Dudley et al. [12], Saadeh et al. [13] and Thorson et al. [14]. Chemoprophylactic anticoagulation is safe after TBI, with a stable head CT 24 h after injury.

Chemoprophylactic anticoagulation also appears to be effective in reducing VTE complications after TBI. In our study, the rate of DVT and PE in the cohort with no routine chemoprophylaxis was 5.6 and 3.7%, while the rate of DVT and PE after routine chemoprophylaxis was 0 and 0.78%, respectively. Similar reductions in VTE rates (from 3 to 1%) were observed by Scudday et al. [9]. Norwood et al. [10] found a 0.83% VTE rate, and Saadeh et al. [13] had 0% VTE. Minshall et al. [8] noted a 0.9% DVT and 1.9% PE rate, and Cothren et al. [11] found a DVT rate of 3.9% and PE rate of 0.8%. Thus, studies compared with historical controls and those with a control group show chemoprophylactic efficacy.

One situation where the role of chemoprophylaxis is less certain is in the TBI patient after emergent/urgent operative neurosurgical intervention. Gerlach et al. [15] treated patients, including 321 TBI patients, 24 h after surgery with nadroparin 0.3 ml/day sc. Post-operative ICH formation was identified in some patients: 2.5% after subdural hematoma evacuation and 1.6% after decompressive craniectomy. No patients with epidural hematoma or surgery for skull fracture developed new ICH. Two patients (0.86%) had VTE. The authors did not compare chemoprophylaxis patients with a cohort who did not receive anticoagulation, leaving the question of increased risk of ICH progression after anticoagulation in TBI unanswered. In our own study, we included post-operative TBI patients without progressive ICH on follow-up head CT 24 h after surgery, but our patient number was too small to make any meaningful conclusions regarding post-operative safety. Chemoprophylaxis has also been shown to be safe after elective neurosurgical procedures [16–18], but generalization to post-operative TBI patients should be made cautiously because of the likelihood of adjacent tissue injury in TBI patients. More data would certainly be helpful in making a stronger recommendation regarding chemoprophylaxis in the post-operative group of TBI patients.

One circumstance where chemoprophylaxis is probably not appropriate is in the setting of coagulopathy associated with head injury. We are not aware of any studies focused on this particular issue, but given the negative impact of coagulopathy on TBI outcome, common sense dictates that chemoprophylaxis should be deferred until coagulopathy has been corrected and not recurred. Consistent with this view, a number of protocols have included coagulopathy as an exclusion criterion for chemoprophylaxis [4,9,10].

Laboratory testing for coagulopathy, including studies such as international normalized ratio, partial thromboplastin time, platelet count and/or thromboelastography, should be completed prior to the initiation of chemoprophylaxis in TBI patients.

In summary, our clinical practice and recommendations to others have been guided by both experience and the clinical data available to us. Granted, most of the data are level III in quality. Even so, for all patients over 15 years old (studies have not adequately addressed pediatric patients), we recommend anticoagulation in the form of enoxaparin 30 mg sc. twice a day (or 0.5 mg/kg if ideal body weight <60 kg) to prevent VTE after TBI for all patients with a stable head CT 24 h after injury or surgery. For those patients with creatinine clearance <30 ml/min or BMI >35, we recommend heparin sc. three times per day. The risk of progression of hemorrhage after a stable 24 h head CT is very low after the institution of chemoprophylaxis. Neurosurgeons and trauma surgeons should be less fearful of consequences of chemoprophylactic anticoagulation, but should remain respectful enough of the potential adverse effects of these agents to monitor TBI patients closely.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending or royalties.

No writing assistance was utilized in the production of this manuscript.