Geerts’ work demonstrated the superiority of 30 mg enoxaparin subcutaneously every 12 h over 5000 units unfractionated heparin at the same interval [3] for VTE prophylaxis after trauma, leading to the widespread acceptance of that regimen. Recently, some have begun to question whether unfractionated heparin 5000 units three times a day may confer protection that is equivalent to enoxaparin [4] while being significantly cheaper. The data on this practice after neurologic injury is limited.

Alternative dosing of enoxaparin has been examined in a limited number of studies. Kurtoglu randomized 120 subjects with TBI or SCI to receive sequential compression devices alone or 40 mg of enoxaparin daily as their sole prophylaxis regimens. While the study showed no significant difference in DVT rates between the groups, there was also an alarming fatal PE rate of 3.3 % in the mechanical prophylaxis arm, and 6.7 % in the enoxaparin arm [5]. Recently, Kopelman examined the effects of increasing the dose of enoxaparin to 40 mg twice a day in a general trauma population. While higher anti-Xa levels were noted, there was no difference in VTE for the 124 subjects in the study [6].

Most studies in of post-TBI prophylaxis utilize enoxaparin as their low molecular weight heparin of choice, but Cothren examined the effect of dalteparin on a subset of 173 TBI patients in 2007 [7]. She found no cases of intracranial hemorrhage progression when dalteparin was started 72 h after injury (albeit with only a 74 % compliance rate). The limited work with other low molecular weight heparins precludes our ability to say much about this option.

An additional complicating factor when discussing the effectiveness of various pharmacologic regimens and doses is the confounding effect of obesity. This was only realized relatively recently, and the data after neurologic injury is limited. Bickford recently showed that an enoxaparin dose of 0.5 mg/kg subcutaneously every 12 h led to target anti-Xa levels in a cohort of obese general trauma patients [8] without an increase in bleeding complications. Given the prevalence of obesity among the American population and the expectation that it will continue to rise, it is likely that this issue will be vetted further in the near future.

Considering the dizzying array of choices and confounders that confront the physician seeking an anticoagulation regimen for their TBI patient, what can be recommended? Our group’s practice has been to use enoxaparin 30 mg subcutaneously every 12 h when initiating prophylaxis after neurologic injury (no disclosures). Our choice was made based on the fact that the preponderance of the evidence shows it to be efficacious and safe. Additionally, our desire to investigate the optimal timing of pharmacologic prophylaxis initiation was aided by a choice of a commonly used regimen (as otherwise the studies of timing would devolve into studies of drug choice). To be sure, new possibilities will continue to arise, some of which will be discussed at the end of this chapter, but for now we feel that the greater need for this field is the elucidation of the timing of initiation rather than the method.

When considering the question of the optimal time to start anticoagulation, we saw a recurring theme through most of the literature on this subject: TBI tended to be treated as a homogenous injury. Study after study just considered TBI as a binary phenomenon in which it was present or absent. While it was understandable that it made for greater methodologic ease in studying the phenomenon, this approach was at variance with what we knew to be our clinical experience with the injury. Seen rightly, TBI is a spectrum of disease in which one could place the patient with a small subarachnoid hemorrhage and a normal Glasgow Coma Score at one end, and the patient with a craniotomy, intracranial monitor, and massive cerebral contusions with midline shift at the other. Why, then, should we treat these patients like they were at the same risk for progression of their injury both prior to and after initiation of prophylaxis?

It was in dealing with this discrepancy that we found the work of Berne and Norwood. These investigators had created an a priori set of injuries which they considered to be candidates for receiving enoxaparin 30 mg subcutaneously every 12 h if a CT scan performed 24 h after injury showed no injury progression [9–11]. These injuries (subdural or epidural hematoma no greater than 8 mm thick at their widest, frontal contusion smaller than 2 cm at its greatest dimension, and a single contusion per lobe) showed no greater rates of growth after anticoagulant initiation than historical controls. We added injury patterns consisting of intraventricular hemorrhage greater than 2 cm in maximum dimension and any degree of traumatic subarachnoid hemorrhage with a negative CT angiogram and in recognition of their work called these the “Modified Berne-Norwood Criteria.” We subsequently used these injuries as a basis for a comprehensive protocol for the timing of anticoagulant initiation [12], and have recently found that these injury patterns predict two tiers of risk for progression [13] (Fig. 22.2). For the lower risk of progression arm, we recommend initiation of enoxaparin at 24 h after injury if a repeat CT scan shows a stable injury pattern. We have performed a pilot randomized trial called the “Delayed vs Early Enoxaparin Prophylaxis [DEEP] Study” showing that the progression rate after starting enoxaparin is (a) low, (b) similar to placebo-treated patients (indicating that the progression appears to be an evolution of disease rather than iatrogenic), and (c) well tolerated as the few patients who progressed were asymptomatic [14].

Those patients who were originally characterized as low risk but progressed on follow up scanning while enoxaparin-naive are upgraded to higher risk for further progression. They are grouped with those patients who present with injuries larger than the Modified Berne-Norwood criteria, and are considered to be candidates for enoxaparin initiation at 72 h after injury if they have radiographically stabilized by that time. Any patients continuing to experience hemorrhage progression at 72 h after injury are off-protocol and have their decisions about initiation tailored to the individual. In general for these rare patients, we delay enoxaparin until they have had 24 h of radiographic stability, whenever that may be. For patients with an intracranial monitor, we will selectively consider them candidates for enoxaparin if they have stable CT scans and smaller hemorrhage patterns. We simply hold the dose of enoxaparin due before pulling the monitor.

We are currently seeking funding for the powered DEEP II study. Much work remains as we must elucidate the effect of these regimens on the DVT rate (since the endpoints studied so far have been toward progression rates to make sure the practice is safe). Additionally, we have gone to a policy of restrictive repeat scanning on patients who present with high GCS [15]. Since the protocol is based around the performance of repeat scanning, the effect has been that we have seen fewer patients being entered into the protocol’s pathway. In reality, this has not mattered as much as would first appear, however, given that most of these high-GCS TBI patients are ambulatory and have short hospital lengths of stay. Those patients who are getting repeat CT scans of the head tend to be more severely injured and consequently in greater need of prophylaxis.

Finally, when looking ahead to the future, we have noted that what most of our pathway seems to generally lead toward is the initiation of enoxaparin at 24 h after the demonstration of radiographic stability whenever that may be. It may be that at the end of our efforts we find that this will be the simple, take-home message.

For the busy clinician who may not have time to sift through large amounts of data, or who wants the imprimatur of national organizations as support for decisions, national guidelines exist for recommendations on the question of VTE prophylaxis after traumatic brain injury . The American College of Chest Physicians (ACCP) promulgates a set of prophylaxis guidelines for all manner of surgical and medical patients leading many to consider this to be the gold standard on the subject [16]. Unfortunately, the most recent recommendations from 2012 for prophylaxis on the route, dose, and timing of pharmacologic prophylaxis after TBI are vague. The recommendation for TBI patients is that sequential compression devices be used until the risk of bleeding is felt to have abated and then to institute pharmacologic prophylaxis with no specification of drug or dose. This recommendation is graded level 2C, indicating a medium level of confidence. The Brain Trauma Foundation has (BTF) put forth its own recommendations, most recently in 2007. Only Level 3 in strength, they are similarly vague. The BTF recommends that TBI patients have some form of mechanical prophylaxis until ambulatory, should receive either unfractionated heparin or low dose heparin in conjunction with mechanical prophylaxis, and that anticoagulation appears to be associated with an increased risk of intracranial bleeding [17]. Finally, the Eastern Association for the Surgery of Trauma (EAST) published guidelines in 2002 which are now quite dated [18]. While they address prophylaxis for all manner of trauma patients, their only statements about TBI are to say that neither unfractionated heparin nor low molecular weight heparin have been adequately studied to make any recommendations.

VTE rates after spinal cord injury are strikingly high, varying from 49 to 100 % in the first 3 months after injury [19]. Providers are well advised to avoid femoral central venous access in patients with spinal cord injury as this can add venous endothelial injury to a patient who is already at high risk by virtue of stasis and hypercoagulability. Interestingly, the risk abates somewhat after about the first two weeks after injury and this must be factored in when creating a prophylaxis strategy.