Associate Professor of Clinical Neurology, Weill Cornell Medical College
Dr. Segal reports no financial relationships relevant to this field of study.
SYNOPSIS: After an exhaustive review of the animal and human studies literature regarding mild traumatic brain injury (mTBI), Haider et al did not reach consistent conclusions regarding evidence for intracranial pressure elevation in human patients who sustain an mTBI.
SOURCE: Haider MN, Leddy JJ, Hinds AL, et al. Intracranial pressure changes after mild traumatic brain injury: A systematic review. Brain Injury 2018; Apr 27. doi: 10.1080/02699052.2018.1469045. [Epub ahead of print].
Traumatic brain injury (TBI) is a term generally used for patients who have suffered a severe blow to the head, resulting in injuries such as cerebral contusion or bleeding into the intracranial vault. Hemorrhage may be multi-compartmental, including the epidural, subdural, subarachnoid, and intra-parenchymal spaces. In addition to the coup and contra-coup effects of direct impact and recoil, acute deceleration of the head leads to shear forces, resulting in diffuse axonal injury (DAI). With DAI, there may be coma and persistent neurological deficits despite normal standard brain imaging (although more advanced techniques, such as tensor diffusion imaging, can show disruption of axonal tracts).
Haider et al aimed to examine the role of intracranial pressure (ICP) in mild traumatic brain injury (mTBI) by performing a systematic review of both animal and human literature. The term “mTBI” is not defined clearly, but closely approximates the term “concussion.” It is defined as a blow to the head associated with loss of consciousness (LOC), amnesia (both retrograde and anterograde), mental status change (e.g., dizziness or confusion), and focal neurological signs that are transient. Additionally, mTBI should have normal brain imaging, with an absence of a visible contusion on computed tomography or magnetic resonance imaging (MRI).
According to the Monro-Kellie doctrine, ICP is made up of three components: brain, blood, and cerebrospinal fluid. Given that the cranium is a closed compartment, when there is an addition to these components, for example, intraparenchymal clot or brain edema, the ICP will rise. Taking the Monro-Kellie doctrine into account, mTBI would increase ICP only if it would produce brain edema below the detection of standard imaging techniques.
Management of mTBI generally is dictated by symptoms, with return to activity based on resolution of headache, dizziness, or other acute symptoms. If elevated ICP plays a role, there would be a crucial need for surrogate markers to detect this.
In their literature review, Haider et al used a series of search terms, including but not limited to: mild traumatic brain injury, concussion, Glasgow Coma Scale 13-15, ICP, intracranial hypertension, brain edema, and papilledema. Their search yielded 1,067 papers, 55 of which passed initial screen and nine of which met inclusion/exclusion criteria. There were five animal studies (primarily rodent) and four human studies, including 95 total subjects, with variable numbers of controls totaling 67. Methodology of human studies included one prospective cohort, one small randomized, clinical trial, and two case-control investigations.
There was no uniformity in the animal models. The largest study involved dropping a weight (averaging 60-120 g) onto the head of a mouse with varying numbers of repetitions (generally daily, for a range of five to 10 days). Another study included sequential impacts on a rat head (helmeted) using a 50 g projectile at a velocity averaging 11.2 m/sec. One animal study involved direct cortical impact on a rodent using a piston following a small craniotomy, producing focal hemorrhage and edema seen post-mortem. Other animal models mimicked a “blast” injury, using pulses of compressed air when the rat was placed in a “shock tube.” Notably, in all of the animal models, the rodent was motionless, with an external force applied, as opposed to much of human head injury, which involves a subject moving at a high velocity and striking an immovable object (e.g., a windshield). Importantly, all of the animal models included direct ICP measurements with invasive probes placed through the cranium. The majority of animal studies included multiple blows. Given this, mTBI can be subcategorized as repetitive mild TBI (rmTBI), which may be a more relevant model than merely an isolated head strike.
Of the five animal studies, two did not show significant ICP increases, but of the three that did, there were consistent data regarding the time course of ICP change. Overall, there was a steady increase in ICP between six and 10 hours post-impact, with a plateau at one day and a trend toward resolution starting at two days and returning to normal or near normal between three and seven days. This ICP trajectory is typical for the edema patterns seen in cortical contusions. The authors suggested that this may explain the “delayed” symptoms in mTBI patients who have delayed symptoms 12-24 hours or more after a blow to the head. Mechanistically, this is unlikely, since humans with concussions (especially those who are initially normal) do not have the types of brain lesions seen in these significantly injured animals. Nonetheless, the importance of sequential exams is important, as return-to-play decisions in athletes are best made after watchful waiting.
The four human studies provided conflicting results without direct ICP measurements available. Two studies using MRI brain volume showed a decrease in MRI volumetrics in mTBI patients compared to controls, which the authors concluded as evidence of decreases in ICP. One study, using phase contrast MRI methodology with assessment of venous compliance, showed significant increases in mTBI patients compared to controls, but this study was done an average of 11 years after impact. The authors cited one clinical study using 3% hypertonic saline showing benefit for mTBI and characterized this as “indirect evidence” of an ICP increase. “Benefit” was decreased as a 2-point decrease on the Faces pain scale in subjects treated with 3% saline compared to subjects treated with normal saline.
This “review” of data on mTBI was not amenable to typical meta-analysis methodology and, therefore, could not reach any statistical conclusions regarding mTBI and ICP. On a descriptive, qualitative level, this review was inconclusive overall. Given the mixture of animal and human data, with variable methods of assessing ICP, and the ambiguity of the nature of mTBI itself, there was too much heterogeneity to reach any consistent conclusion.
In the animal models, there was little uniformity in the mechanism used for brain trauma, with the overall severity likely more intense than in human concussion (including a direct piston blow to the cortex following craniotomy). The authors acknowledged that assessment for mild ICP changes in humans could be made only using noninvasive means. Lower brain volumes were seen in head-injured patients compared to controls, but it is unclear how this was related to ICP. More accurate, complex MRI analysis using phase contrast and venography techniques may provide useful data regarding ICP, but the only data were in patients who were years removed from their injury and is of questionable significance. One possible noninvasive surrogate marker of ICP elevations in critically ill subjects is optic nerve sheath diameter as measured by trans-orbital transcranial Doppler. This technique may be useful in detecting ICP increases in mTBI, but may not be sensitive enough for subtle increases in ICP, with a high dependency on the specific diameter cutoff values chosen.
Although these authors did not reach conclusions regarding ICP in mTBI, it is important work, given concerns that repetitive episodes of mTBI (concussion) can lead to accelerated brain degeneration (chronic traumatic encephalopathy). This review draws attention to the significant challenges that remain in understanding of how such mTBI injuries may damage the brain over time.