Novel Focal Therapeutic Hypothermia Device for Treatment of Acute Neurologic Injury: Large Animal Safety and Efficacy Trial

Abstract

Objective  Therapeutic hypothermia is a potentially powerful and controversial clinical tool for neuroprotection following acute neurologic pathology, particularly vascular injury. Indeed, therapeutic hypothermia remains a standard of care for postcardiac arrest ischemia and acute neonatal hypoxic-ischemic encephalopathy, improving both survival and outcomes. Although therapeutic hypothermia remains promising for cellular and systems-based neuronal protection in other neurologic injury states, the systemic side effects have limited clinical utility, confounded analysis of potential neurologic benefits, and precluded the completion of meaningful clinical trials.

Methods  To address such limitations, we developed and tested a novel, minimally invasive, neurocritical care device that employs continuous circulation of cold saline through the pharyngeal region to deliver focal cerebrovascular cooling. We conducted a preclinical safety and efficacy trial in six adult porcine animals to assess the validity and functionality of the NeuroSave device, and assess cooling potential following middle cerebral artery occlusion ( n  = 2).

Results  NeuroSave consistently lowered brain parenchymal temperature by a median of 9°C relative to core temperature within 60 minutes of initiation, including in ischemic cerebral parenchyma. The core body temperature experienced a maximal reduction of 2°C, or 5% of body temperature, with no associated adverse effects identified.

Conclusion  The present study uses a large animal preclinical model to demonstrate the safety and efficacy of a novel, noninvasive device for the induction of robust and systemically safe hypothermia within the brain.

Introduction

Therapeutic hypothermia is an established clinical therapy for anoxic injury following cardiac arrest and hypoxic-ischemia neonatal encephalopathy, and has shown tremendous preclinical success for inducing neuroprotection following vascular or traumatic neurologic injury. ^{1 2} However, equivocal and often conflicting clinical trial results have made conclusive formal recommendations regarding hypothermia utility for other acute neurologic injury states challenging. ^{3 4 5} Many of these challenges stem from highly morbid systemic side effects unrelated to cooling of the central nervous system (CNS), including infections, coagulopathies, cardiac arrhythmias, and pulmonary edema, among others. In addition to the numerous adverse systemic effects, the time required for attaining total body cooling is considerable and likely further limits usefulness for treating hyperacute disease and remains an issue for currently approved conditions. ⁶

Ischemic infarctions in the proximal middle cerebral artery (MCA) may lead to mortality rates as high as 80% as a result of malignant cerebral edema and raised intracranial pressure (ICP). ^{7 8} Advanced in medical and endovascular treatments for acute proximal ischemic disease have radically altered the clinical management landscape, with an even greater proportion of patients going through these stroke pipelines. However, even with maximal procedural and medical interventions, many acute stroke patients experience residual ischemic penumbra and resultant cerebral edema leading to considerable short- and long-term impacts on morbidity and quality of life outcomes. Clinical trials evaluating systemic hypothermia for acute proximal MCA occlusions were equivocal, and often short-lived, due to the severity of systemic effects; however, these trials may also have been confounded by including patients for cooling regardless of reperfusion status. ^{9 10} The Reperfusion and Cooling in Cerebral Acute Ischemia (ReCCLAIM) I trial, a safety and feasibility trial, did show relative safety and a small protective effect against intracerebral hemorrhage. ¹¹ Therefore, it remains difficult to assess the clinical impact that targeted hypothermia may have in the CNS for acute ischemic events. Furthermore, targeted cooling may also be invaluable for attaining normothermia during acute severe febrile illness. Aggressive prevention of hyperthermia has been shown to reduce total infarct areas by up to 40 to 50% in MCA occlusions and improved the efficacy in reperfusion studies on infarct volumes and total damaged brain volumes in rat models. ¹² Hypothermia is likely to have a wide array of clinical applications, in particular acute ischemic events, when the systemic side effects are ameliorated. The long-term goal of this research is to improve neurologic outcomes following acute neurologic injury, particularly vascular insults. This is to be achieved through development of a relatively simple device that promotes rapid and effective neuroprotection through focal CNS cooling during states of acute injury—when the risk of permanent neurologic harm is most severe. Our study evaluated the NeuroSave device for obtaining safe, rapid, and reliable hypothermia limited to the brain.

Materials and Methods

Study Design

Adult domestic male and female porcine animals, weighing 50 kg, were utilized to assess the safety and efficacy of a novel device for inducing focal therapeutic hypothermia within brain. Adult pigs were chosen due to well-characterized shared anatomic and physiologic characteristics with humans, in particular, similarly sized bloody supply distributions and comparable brain sizes. A power analysis was performed to verify that a power > 0.80 was attained while reducing redundant animal usage. Data collection was initiated throughout application of the NeuroSave device, and completed once the animals returned to precooling baselines.

Animal Studies

All experiments and animal housing were conducted according to the National Institutes of Health (NIH, Bethesda, Maryland, United States) guidelines for animal care and safety, with the approval and under the auspices of the Mayo Clinic Institutional Animal Care and Use Committee (IACUC). The animals were anesthetized with intramuscular injection of Telazol and xylazine. An ear vein catheter was placed in each ear and the animal was intubated. Once in the procedure room, the animal was placed on isoflurane carried by 100% oxygen. The body core temperature of the animal was monitored by placement an inferior vena cava (IVC) and rectal temperature probe.

Under anesthesia, a burr hole was made to access the brain parenchyma. After opening the dura, a 0.5-mm-diameter fiber optic temperature probe was inserted 2 cm in the cerebrum. Two animals underwent additional right-sided frontotemporal craniectomies. On the right, after opening the arachnoid to release cerebrospinal fluid, the head of the bed was elevated to provide maximal brain relaxation. Under direct high surgical magnification, bipolar electrocautery was used to occlude both right MCA branches running beneath the posterior frontal lobe, in one animal, a fiber optic temperature probe was also placed in the infarction penumbra. ^{13 14}

Next, the animal was turned supine, and a venous cut down was performed over the left common femoral vein. A 7-French sheath was placed into the femoral vein. A 7-French Envoy guiding catheter was also placed with a 0.035-inch wire into the IVC. After confirmation of correct placement was verified via fluoroscopy, a fiber optic temperature sensor was threaded into the IVC. Then, the NeuroSave nasal mask was placed to establish a fluid seal on each nostril for fluid delivery. The NeuroSave esophagus catheter (balloon cuffed multilumen) was placed through the mouth and esophagus until the distal end was in the stomach. The proper placement of the catheter's gastric balloon was confirmed using fluoroscopy. Following visual verification, the balloon was inflated with 250 mL of air to an approximate diameter of 8 cm. Subsequently, a catheter was placed under mild tension to create a fluid-tight seal at the lower esophageal junction and to prevent fluid from entering the stomach. A noncuffed oral suction tube was placed into the oral cavity for fluid recovery.

The animals were then taken for dual-energy computed tomography (CT) imaging, including CT perfusion. Baseline CT scans were performed to assess the presence of cerebral infarction. The heater/cooler was activated to circulate 37°C fluid to one side of the heat exchanger. The NeuroSave device was connected to a saline circuit, and the system was primed while isolated from the animal-contacting catheters, and circulation started to bring the circulating saline to near 37°C. Flow rate was set to 3 L/min. Once a baseline was established, the heater/chiller was set to 2°C to begin cooling. Continuous fiber optic temperature probe recording was performed throughout the duration of pre- and postcooling. Cerebral cortex temperature >10°C colder than baseline and <10°C change in the whole body temperature will be considered as successful cooling of the brain after activation of NeuroSave device for 60 minutes. After 60 minutes of cooling, a steady-state brain/body differential was established and a step-wise rewarming protocol was initiated. The heater/chiller temperature was adjusted to 13°C for 15 minutes, then 24°C for 15 minutes, with continued step-wise rewarming until rewarming was completed, with monitoring halted after 150 minutes. Two animals underwent additional CT imaging of the brain, chest, and upper abdominal regions to evaluate for any signs of tissue injury. Following completion of the imaging and confirmation of full body rewarming, the catheters were removed and the animals were euthanized in accordance with IACUC guidelines. Tissue samples of the brain, pharyngeal mucosa, esophagus, and lungs were harvested for histopathologic analysis. The postcooling imaging was performed on non-MCA occluded animals ( n  = 2), while the gross and histopathologic assessment took place on all experimental animals. Statistical analyses were completed using JMP statistical software (v14). Normally distributed data (D'Agostino and Pearson's omnibus normality test) was assessed using a two-tailed t -test. Nonnormally distributed data were analyzed using the nonparametric Mann–Whitney's U -test.

Results

Six adult domestic pigs underwent targeted hypothermia with the NeuroSave device, five males and one female ( Fig. 1 ). Study animals underwent continuous monitoring of rectal region, pharyngeal space, IVC, and cerebral temperatures prior to and during initiation of hypothermia, as well as throughout the rewarming phase. The median cerebral temperature after 60 minutes of targeted cooling was 27.8°C (range: 26.7–29.4°C) ( Fig. 2A, B ), with all animals returning to within 1°C of their baseline temperature by 70 minutes (range: 55–105 minutes). The median temperatures for the pharyngeal space and IVC at 60 minutes were 5.6°C (range: 4.6–11.2°C) and 36.5°C (range: 35.9–37.1°C), respectively ( Fig. 2B ). All animals returned to cerebral normothermia within 90 minutes of trial cessation (median change from baseline 0.3°C, range: 0.1–0.8; p  = 0.52). Throughout the duration of continuous monitoring and therapeutic hypothermia, the core body temperatures never fell by more than 2°C from baseline recordings. CT scans generated during cooling and postmortem tissue analyses, as well as gross histopathologic assessment, demonstrated no findings of pharyngeal, cardiovascular, or pulmonary injury. CT perfusion studies performed throughout the hypothermic process demonstrated a decrease in cerebral blood flow during observed maximal hypothermia, with associated increase in mean transit time and stable overall cerebral blood volume ( Fig. 3A–C ).

Fig. 1.

NeuroSave device components. NeuroSave cold saline circulating system component schematic for human use ( A ) and representative image of the focal hypothermia device while in use in the adult pig model ( B ).

Fig. 2.

Successful focal cerebral hypothermia from NeuroSave. Fiber optic temperature probes (°C) in the brain recorded throughout initiation, cooling, and rewarming of all six pigs (P1–P6) in the trial ( A ). Median temperatures of the brain parenchyma, IVC, and pharynx/snout throughout the process of therapeutic hypothermia using the NeuroSave device ( B ). Fiber optic temperature probe recordings placed within MCA-occluded ischemic parenchyma versus the contralateral nonischemic hemisphere during the hypothermic cycle ( C ). IVC, inferior vena cava; MCA, middle cerebral artery.

Fig. 3.

Cerebral vasoreactivity is maintained during hypothermia. CT perfusion studies during baseline, initiation of hypothermia, maximal observed hypothermia, rewarming, and return to normothermia postrearming highlighting changes in cerebral blood flow ( A , mL/100 mL/min), cerebral blood volume ( B , mL/100 mL), and mean cerebral blood transit times ( C , second). CT, computed tomography.

Two adult pigs underwent open surgical occlusion of the right MCA territory blood supply. Because of the rete mirabile in the pig, this included occlusion of the two main proximal MCA branches. Postmortem histopathologic staining with triphenyl tetrazolium chloride, a marker for metabolically inactive tissue, and corresponding volumetric analysis demonstrated ∼50% ischemic tissue on the ipsilateral hemisphere ( Fig. 4A–F ). One animal was allowed an extended 2 hours for the ischemia to progress before performing histopathology, which allowed further progression of tissue ischemia. Intraparenchymal fiber optic temperature probes were surgically placed in the ipsilateral penumbra, or in the control contralateral side to monitor whether the reduced perfusion prevents symmetric cooling to the ischemia zone. There was no statistical difference in the median parenchymal temperature during the hypothermic period between ischemic and nonischemic brain parenchyma (31.7 vs. 30.6°C, p  = 0.09) ( Fig. 2C ).

Fig. 4.

Histopathology following MCA occlusion. Pathologic images of two representative pig brains following right MCA occlusion showing early ischemic changes ( A, B ). Coronal histopathologic sectioning of pig brain following MCA occlusion and TTC staining confirming MCA-distribution ischemia ( C ). Representative individual brain slices following TTC staining further illustrating acute MCA territory ischemic changes ( D–F ). MCA, middle cerebral artery; TTC, tetrazolium chloride.

Discussion

Hypothermia-mediated neuroprotection is proposed to act on multiple cell intrinsic and system-wide metabolic mechanisms—in both acute and chronic phases. A well-described reduction in cerebral metabolism, via oxygen consumption and adenosine triphosphate modulation, as well as preservation of cerebral glucose and blunted lactate production highlights acute protective effects. Interestingly, cerebral blood flow also seems to be linearly coupled with temperature, which may indicate preservation in vasoreactivity even in hypothermic states. ^{15 16} In fact, cerebral metabolic rates have been shown to decrease by as much as 7% for every 1°C drop in body temperature in mammalian studies ¹⁷ —meaning NeuroSave has the potential to reduce metabolic demand by up to 60% and improve restoration of physiologic energetic parameters during reperfusion for ischemic pathologies.

In the more chronic phase, hypothermia has been implicated in suppressing injurious inflammatory and apoptotic responses, reduced cerebral edema, and may even positively impact neuro- and gliogenesis. ^{2 15 18} Presumably, some of the neuroprotective effects from hypothermia in ischemic disease are the reduction in cerebral edema, preservation of microvascular flow, and reduction in degree of blood–brain barrier disruption. ¹⁹ A correlative decrease in ICP has been observed in clinical trials with the induction of hypothermia, and is likely also attained with the NeuroSave device. Therefore, it is probable that targeted CNS hypothermia will be beneficial for other neurologic conditions with similar underlying pathologies, such as traumatic brain injuries, aneurysmal and traumatic subarachnoid hemorrhage, and intraoperatively during mass lesion resections, among other types of acute neurologic injury that require both neurologic protection and reduction of cerebral edema and ICP.

One particular acute neurologic injury where the NeuroSave device may be germane is acute proximal MCA occlusion. As rapid restoration of flow to proximal occlusions in the brain continues to advance as a major health care initiative, the periacute medical management of these patients will become even more important—as older and sicker patients will survive the initial insult of their disease with varying risks for subsequent neurologic decline and injury. Of particular note, older patients who do not qualify for certain surgical interventions to decompress acutely elevated ICPs secondary to MCA infarcts may benefit from other strategies to lower ICP and preserve precious neurologic function. ²⁰ Therefore, it is essential for the development of effective clinical tools for optimizing these patients medically to achieve desired outcomes. Systemic hypothermia was evaluated in this particular patient population, those >60 years of age and >2/3 MCA territory infarction. Unfortunately, the trial was limited by the frequent of adverse systemic events: electrolyte imbalances (82%), cardiac arrhythmias (45%), bradycardia (45%), pneumonia (55%), pulmonary edema (27%), bleed diathesis (45%), and death (18%). ²¹ Other trials, evaluating systemic hypothermia combined with thrombolysis in all types of ischemic strokes identified again relatively high degrees of cardiac dysfunction (80%), infections (40%), and mortality (30%). Endovascular hypothermia for cerebral ischemia showed tremendous difficulties obtaining rapid, reliable cooling. In other trials, up to 60% of the cohort failed to meet the target temperatures and in another, the rate of cooling was around 1°C per hour. ^{10 22 23 24} Interestingly, patients who did obtain hypothermic targets did show reduced cerebral edema and ICP, and trended toward significance in volume reduction of diffusion-weighted imaging ischemic areas ^{9 24} —indicating that therapeutic hypothermia in humans is likely to address the desired pathophysiologic targets to provided clinical benefit, provided the systemic events can be significantly reduced. The ReCCLAIM I trial, a feasibility and safety trial that required definitive reperfusion for study inclusion, showed benefit for reduced rates of intracerebral hemorrhage in the intervention arm. ¹¹ This finding suggests that therapeutic cooling may not only benefit long-term neurologic preservation but also reduce adverse neurologic events in the acute setting.

In addition to the systemic adverse effects, systemic hypothermia also necessitates considerable time to attain desired degrees of cooling, and likely limits the utility of therapeutic systemic hypothermia to less hyperacute states. Furthermore, many patients suffering neurologic injury, notably traumatic brain injuries and acute ischemic disease, also develop hyperthermia, which carries important implications on neurologic morbidity and mortality and may benefit from rapid, targeted cooling ^{25 26} although therapeutic hypothermia for traumatic brain injuries may not provide benefit. ²⁷ Alternatively, NeuroSave also has the theoretical capacity for inducing selective CNS hyperthermia—if warmed instead of cooled saline if infused into the pharyngeal space. While largely theoretical and in early preclinical stages, there is evidence to suggest that targeted hyperthermia may conversely activate the immune surveillance and aide in leukocyte trafficking, the activation of dendritic cells, and induction of heat shock protein release—to the benefit of augmenting oncologic therapies, particularly as an adjuvant to radiation and/or immunomodulation. ^{28 29}

The optimal rate and timing of rewarming after therapeutic hypothermia remains unknown, and likely has a different set of parameters for unique pathologies. Clinical trials have identified difficulties with rebound elevated ICP and hemorrhagic transformation as major issues related with rewarming, ^{9 30} although Hong et al showed a reduction in rates of rewarming-associated hemorrhagic transformation with a prolonged warming course. ³⁰ This study demonstrated that controlled rewarming can be accomplished with the NeuroSave device; however, we did not address appropriate clinical timing of rewarming as the animals were killed for histopathologic analysis once the study was complete and rewarming remains a relevant consideration for future studies. One additional caveat to the implementation of the NeuroSave device clinically is the requirement for patients to be intubated—presumably with a degree of sedation, making frequent neurologic examinations difficult. Unfortunately, surface cooling, which often poorly tolerated by awake patients, is inefficient and unreliable at inducing rapid targeted cooling. It is likely NeuroSave would be reserved for cases of severe neurologic injury, where intubation may be required as a sequela of injury, or for cases where a brief period of sedation may be tolerated.

Conclusion

Development of an effective neuroprotective agent would be a powerful tool to combat significant morbidity and mortality associated with traumatic and acute ischemic neurologic injuries. Therapeutic hypothermia has long been heralded as an option; however, slow and highly morbid systemic cooling has proven unreliable. Alternatively, targeted CNS hypothermia, through convective cooling of cervical arteries leads to rapid, highly controlled cerebral hypothermia without systemic side effects would be a powerful clinical tool for an array of acute neurologic injury states. Preclinical testing of the NeuroSave device demonstrates safe, precise, therapeutic hypothermia throughout the porcine brain, even in regions of acute ischemia.

Funding Statement

Funding The study was funded by Mayo Clinic Radiology Research Committee.