Analysis of the dynamics of experimental epidural bleeding in swine

Development of a Novel Epidural Hemorrhage Model in Swine

INTRODUCTION

Traumatic brain injury is a major public health concern. Among patients with severe traumatic brain injury, epidural hemorrhage is known to swiftly lead to brain herniation and death unless there is emergent neurosurgical intervention. However, immediate neurosurgeon availability is frequently a problem outside of level I trauma centers. In this context, the authors desired to test a novel device for the emergent management of life-threatening epidural hemorrhage. A review of existing animal models determined that all were inadequate for this purpose, as they were found to be either inappropriate or obsolete. Here, we describe the development of a new epidural hemorrhage model in swine (Sus scrofa, 18-26 kg) ideal for translational device testing.

MATERIALS AND METHODS

Vascular access was achieved using an ultrasound-guided percutaneous Seldinger catheter-over-wire technique with 5 Fr catheters placed in the bilateral carotid arteries, for continuous blood pressure and to allow for withdrawal of blood for creation of epidural hemorrhage. To simulate an actively bleeding and life-threatening epidural hemorrhage, unadulterated autologous blood was infused from the vascular access point into the epidural space. To be useful for this application and clinical scenario, brain death needed to occur after the planned intervention time but before the end of the protocol period (if no intervention took place). An iterative approach to model development determined that this could be achieved with an initial infusion rate of 1.0 mL/min, slowed to 0.5 mL/min after the first 10 minutes, paired with an intervention time at 15 minutes. All experiments were performed at Oregon Health & Science University, an Association for Assessment and Accreditation of Laboratory Animal Care accredited facility. Oregon Health & Science University's Institutional Animal Care and Use Committee, as well as the United States Army Animal Care and Use Review Office, reviewed and approved this protocol before the initiation of experiments (respectively, protocol numbers IP00002901 and 18116010.e001).

RESULTS

The final developed model allows for the infusion of a known volume of autologous, unadulterated blood directly into the epidural space, without the use of a balloon or other restricting membranes, and is rapidly fatal in the absence of intervention.

CONCLUSIONS

This animal model is the first to mirror the expected clinical course of epidural hemorrhage in a physiologically relevant manner, while allowing translational testing of emergency devices. This model successfully allowed the initial testing of a novel interventional device for the emergent management of epidural hemorrhage that was designed for use in the absence of traditional neurosurgical capabilities.

Contrast Media Leakage to the Cerebrospinal Fluid after Intravenous Injection

We have tested a method to evaluate leakage of i.v. injected contrast media (CM) through the 3 partitions between blood and cerebrospinal fluid (CSF) in controls and in patients with acute cerebrovascular disease (ACBVD) to detect differences between normal brains and brains with ischemic lesions. High-osmolar (HOCM) and low-osmolar (LOCM) CM were used. In 55 patients and in 41 controls who underwent CT after i.v. contrast administration, lumbar CSF was collected 1 hour after injection and the iodine content in the CSF was measured. The concentration of iodine in CSF was very low, between 0.57 and 11.20 ng/l, and no significant difference could be found between patients and controls or between HOCM and LOCM. We conclude that under the conditions mentioned above, iodine detected in the human lumbar CSF does not reflect the true leakage of contrast agent through the blood-brain barrier.

Changes in Intracranial Morphology, Regional Cerebral Water Content and Vital Physiological Variables during Epidural Bleeding

Epidural bleeding was produced in 8 anaesthetised and heparinised dogs by an artificial system. Changes in vital physiological variables were related to intracranial shifts and tissue water content assessed with MR imaging. Six animals survived while 2 succumbed. In the surviving animals intracranial shifts and compressions remained unchanged from an early stage. The cerebral perfusion pressure was reduced from between 80 and 110 mm Hg to between 40 and 60 mm Hg. Some increase in supratentorial white matter tissue water was observed. In the lethal experiments cerebral perfusion pressure fell to less than 40 mm Hg. Moreover, secondary delayed anatomical changes were seen including hydrocephalus. Increase in cerebral tissue water was more intense and widespread than in the survivors. These findings indicate that the outcome of epidural bleeding is related to cerebral perfusion pressure with secondary deterioration resulting from additional volume loading from increased tissue water and hydrocephalus.

Pre-Clinical Traumatic Brain Injury Common Data Elements: Toward a Common Language Across Laboratories

Traumatic brain injury (TBI) is a major public health issue exacting a substantial personal and economic burden globally. With the advent of "big data" approaches to understanding complex systems, there is the potential to greatly accelerate knowledge about mechanisms of injury and how to detect and modify them to improve patient outcomes. High quality, well-defined data are critical to the success of bioinformatics platforms, and a data dictionary of "common data elements" (CDEs), as well as "unique data elements" has been created for clinical TBI research. There is no data dictionary, however, for preclinical TBI research despite similar opportunities to accelerate knowledge. To address this gap, a committee of experts was tasked with creating a defined set of data elements to further collaboration across laboratories and enable the merging of data for meta-analysis. The CDEs were subdivided into a Core module for data elements relevant to most, if not all, studies, and Injury-Model-Specific modules for non-generalizable data elements. The purpose of this article is to provide both an overview of TBI models and the CDEs pertinent to these models to facilitate a common language for preclinical TBI research.

Animal models of traumatic brain injury

Traumatic brain injury (TBI) is a major health issue comprising a heterogeneous and complex array of pathologies. Over the last several decades, numerous animal models have been developed to address the diverse nature of human TBI. The clinical relevance of these models has been a major point of reflection given the poor translation of pharmacologic TBI interventions to the clinic. While previously characterized broadly as either focal or diffuse, this classification is falling out of favor with increased awareness of the overlap in pathologic outcomes between models and an emerging consensus that no one model is sufficient. Moreover, an appreciation of injury biomechanics is essential in recapitulating and interpreting the spectrum of TBI neuropathology observed in various established models of dynamic closed-head TBI. While these models have replicated many specific features of human TBI, an enhanced context with clinical relevancy will facilitate the further elucidation of the mechanisms and treatment of injury.

The lucid interval associated with epidural bleeding: evolving understanding

The aim of this paper was to elucidate the evolution of our understanding of the term "lucid interval." A number of texts were reviewed to assess their suitability for analysis. The primary requirement was that the text contain detailed descriptions of a series of patients. Details of the clinical course, the findings and timing of surgery, and, when relevant, the time of death and postmortem findings were required. Books written by Henri-François Le Dran, Percival Pott, and James Hill fulfilled these criteria. Surgical findings included the presence and type of fractures, changes in the bone, separation of periosteum, malodorous or purulent material, tense brain, and hematoma. Postmortem findings supplemented and/or complemented the surgical findings. The courses of the patients were then tabulated, and the correlation between different clinical and operative findings was thereby determined. Our understanding of a lucid interval began in the early 18th century with the work of Henri-François Le Dran and Percival Pott in London. They did not, however, demonstrate an interval without symptoms between trauma and deterioration in patients with epidural hematomas (EDHs). The interval they described was longer than usually expected with EDHs and occurred exclusively in patients who had a posttraumatic infection. In 1751, James Hill, from Dumfries, Scotland, described the first hematoma-related lucid interval in a patient with a subdural hematoma. The first case of a lucid interval associated with an EDH was described by John Abernethy. In the 19th century, Jonathan Hutchinson and Walter Jacobson described the interval as it is known today, in cases of EDH. The most recent work on the topic came from studies in Cincinnati and Oslo, where it was demonstrated that bleeding can separate dura mater and that hemorrhage into the epidural space can be shunted out via the veins. This shunting could delay the accumulation of a hematoma and thus the rise in intracranial pressure, which in turn would delay the development of symptoms. The lucid interval as previously conceived was not properly understood by the French school or by Percival Pott and Benjamin Bell, who all described a symptom-free period prior to the development of infection. The first to have a proper understanding of the interval in relation to an EDH was John Abernethy. The modern description and definition of the lucid interval was the work of Hutchinson and Jacobson in the latter half of the 19th century. Understanding of the pathophysiology of the lucid interval has been advanced by the work of Ford and McLaurin in Cincinnati and a group in Oslo, with the demonstration of what it takes to loosen dura and how an arteriovenous shunt slows down for a while the accumulation of an EDH.

Regional blood flow in brain and peripheral tissues during acute experimental arterial subdural bleeding

The effects of a large intracranial arterial subdural bleeding on regional blood flow in the brain (rCBF) and in other body organs were studied, using a porcine model. The bleeding was produced by leading blood through a catheter from the abdominal aorta via an electronic drop recorder into the subdural compartment (SDC) over the left cerebral hemisphere. Pressures in the right lateral cerebral ventricle and in the cisterna magna were recorded along with 15 other vital parameters. Measurements of rCBF were carried out using radioactive microspheres 1) before the start of bleeding, 2) during the early bleeding phase, and 3) during the late bleeding phase. When the bleeding was initiated, the intracranial pressures rose within one minute to a level approximately 40 mmHg below the systemic arterial pressure, whilst the latter usually decreased 30-40 mmHg. In the subsequent early bleeding phase the cerebral perfusion pressure and the bleeding pressure fluctuated at a level of approximately 40 mmHg for several minutes. In the late bleeding phase, the perfusion pressure decreased maximally, even when a Cushing reaction was activated. During the early bleeding phase the changes in rCBF varied between the cerebral regions. However, the mean flow remained largely constant in the presence of a decreasing cerebrovascular resistance, indicating that autoregulation of CBF was intact. Concomitantly, cardiac output and heart rate decreased, whilst regional blood flow in extracerebral organs tended to increase, possibly due to an intracranial effect on the autonomic nervous system. In the late bleeding phase, rCBF was critically reduced in all regions, in spite of a marked rise in systemic arterial pressure.(ABSTRACT TRUNCATED AT 250 WORDS)

A quantitative study of some factors affecting the outcome of experimental epidural bleeding in swine

During an experimentally induced aggressive epidural bleed the effect on outcome of haematoma volume, cerebral perfusion pressures, intracranial pressure gradients and ventilation were examined in a swine model. Two groups of experiments were performed using either spontaneous ventilation (group 1, n = 6) or mechanical ventilation for 1 hour (group 2, n = 7). The preparations were otherwise identical. An animal was considered to have succumbed when the EEG became irreversibly isoelectric within a total follow-up time of 80 minutes. Mechanical ventilation had a marked effect on survival. All spontaneously ventilated animals succumbed, 4 of them in less than 60 minutes, the remaining 2 between 60 and 80 minutes after the start of bleeding. All mechanically ventilated animals survived for the 60 minutes while the ventilator was connected. Following disconnection 2 animals started to breathe spontaneously and survived the final 20 minutes of the 80 minutes of the follow-up time. The remaining 5 succumbed following apnoea. The size of haematoma did not differ significantly between the groups. Two additional factors, hypoventilation and a secondary rise in supratentorial pressure, contributed to a lethal outcome. Hypoventilation was an inevitable precursor of the isoelectric EEG. There was a close correlation between the development of hypoventilation and intracranial herniation. A secondary rise in supratentorial pressure, unrelated to ventilation, was seen after cessation of bleeding in 8/13 cases. It was associated with a falling supratentorial perfusion pressure and EEG attenuation, suggesting a secondary intracranial expansion, possibly due to oedema, hydrocephalus or both. It is concluded that mechanical ventilation in the acute stage of epidural bleeding may be of clinical value.(ABSTRACT TRUNCATED AT 250 WORDS)

Cerebral blood flow during experimental epidural bleeding in swine

Regional cerebral blood flow (rCBF) was studied during an aggressive epidural bleed, using a ventilated swine model. rCBF, regional organ blood flow and cardiac output were measured using the radioactive microsphere technique. Blood flows were measured prior to the start of bleeding (Stage 1), when intracranial pressures had reached a plateau and supratentorial perfusion pressure was reduced by about 50% (Stage 2), and at isoelectric EEG (Stage 3). Supratentorial rCBF did not change significantly between stages 1 and 2 while rCVR decreased, implying autoregulatory activity. Cerebral ischaemia developed between stages 2 and 3 when rCBF values fell to levels between 20 and 50% of control values. Infratentorial rCBF changes were similar but less marked, so that adequate brain stem perfusion was maintained below the upper mesencephalon. The left temporal and left parietal cortex and upper mesencephalon suffered a greater reduction in rCBF than other regions, due to proximity to the haematoma and tentorial herniation. The supratentorial perfusion pressure at stage 2 was 60 mm Hg associated with a haematoma volume of 6% of the intracranial volume (ICV). The infratentorial perfusion pressure never fell below 60 mm Hg. The Cushing response was absent when the EEG became isoelectric. This is tentatively ascribed to the absence of hypoxia, because mechanical ventilation was used. Instead systemic arterial hypotension accompanied bleeding in this ventilated model. This hypotension was due to falling cardiac output and peripheral vasodilation.