Trigeminal Nerve Stimulation Improves Cerebral Macrocirculation and Microcirculation After Subarachnoid Hemorrhage: An Exploratory Study

BACKGROUND:

Delayed cerebral ischemia (DCI) is the most consequential secondary insult after aneurysmal subarachnoid hemorrhage (SAH). It is a multifactorial process caused by a combination of large artery vasospasm and microcirculatory dysregulation. Despite numerous efforts, no effective therapeutic strategies are available to prevent DCI. The trigeminal nerve richly innervates cerebral blood vessels and releases a host of vasoactive agents upon stimulation. As such, electrical trigeminal nerve stimulation (TNS) has the capability of enhancing cerebral circulation.

OBJECTIVE:

To determine whether TNS can restore impaired cerebral macrocirculation and microcirculation in an experimental rat model of SAH.

METHODS:

The animals were randomly assigned to sham-operated, SAH-control, and SAH-TNS groups. SAH was induced by endovascular perforation on Day 0, followed by KCl-induced cortical spreading depolarization on day 1, and sample collection on day 2. TNS was delivered on day 1. Multiple end points were assessed including cerebral vasospasm, microvascular spasm, microthrombosis, calcitonin gene-related peptide and intercellular adhesion molecule-1 concentrations, degree of cerebral ischemia and apoptosis, and neurobehavioral outcomes.

RESULTS:

SAH resulted in significant vasoconstriction in both major cerebral vessels and cortical pial arterioles. Compared with the SAH-control group, TNS increased lumen diameters of the internal carotid artery, middle cerebral artery, and anterior cerebral artery, and decreased pial arteriolar wall thickness. Additionally, TNS increased cerebrospinal fluid calcitonin gene-related peptide levels, and decreased cortical intercellular adhesion molecule-1 expression, parenchymal microthrombi formation, ischemia-induced hypoxic injury, cellular apoptosis, and neurobehavioral deficits.

CONCLUSION:

Our results suggest that TNS can enhance cerebral circulation at multiple levels, lessen the impact of cerebral ischemia, and ameliorate the consequences of DCI after SAH.

ABBREVIATIONS:

BA

basilar artery

CGRP

calcitonin gene-related peptide

CSD

cortical spreading depolarization

CV

cerebral vasospasm

DCI

delayed cerebral ischemia

DIND

delayed ischemic neurological deficits

HIF-1α

hypoxia inducible factor 1 subunit alpha

HPF

high powered field

ICAM-1

intercellular adhesion molecule 1

TNS

trigeminal nerve stimulation

TUNEL

tdt-mediated deoxyuridine triphosphate nick-end labeling

Tx

trigeminal neurectomy.

Aneurysmal subarachnoid hemorrhage (SAH) is a complex cerebrovascular disease with high morbidity and mortality due, in part, to delayed cerebral ischemia (DCI).¹ Despite tremendous efforts, few interventions have been shown to significantly improve outcomes in these patients.² This may be because of the fact that most therapies investigated have targeted single pathophysiological mechanisms, whereas both experimental and clinical evidence indicate that DCI is a multifactorial process.³ Tissue-level hemodynamic imaging has demonstrated a discordance between capillary level perfusion and large artery vasospasm in patients with SAH, with experimental evidence suggesting microvascular compromise in SAH.^4–11 It is increasingly recognized that cerebral microcirculatory changes play an important role in causing microvascular spasm^12,13 and microthrombosis formation,¹⁴ ultimately leading to DCI after SAH. Novel interventions that can modulate both cerebral macrocirculation and microcirculation may interrupt the pathophysiological mechanisms underlying DCI.

The trigeminal nerve richly innervates the cerebral blood vessels and maintains connections to sympathomimetic and cardiovagal nuclei, allowing it to dramatically alter neurovascular homeostasis.¹⁵ Importantly, the trigeminal nerve is able to affect large cerebral arteries, arterioles, and capillaries via direct axonal projections to vessel walls and centrally mediated mechanisms through brainstem nuclei. Upon stimulation, the trigeminal nerve modulates cerebral blood flow (CBF) and produces profound vasodilation by releasing a host of vasoactive agents, such as calcitonin gene-related peptide (CGRP).¹⁶ Prior research has demonstrated that trigeminal nerve stimulation (TNS) can alleviate dysregulation of CBF and resolve large artery cerebral vasospasm (CV) after experimental SAH.^17–19 However, whether TNS can improve cerebral microcirculation, mitigate DCI, and improve neurological outcomes has yet to be investigated.

In this study, we sought to investigate whether TNS can restore impaired cerebral macrocirculation and microcirculation after SAH. We established a clinically relevant animal model, with cortical spreading depolarization (CSD) as a “second hit” to induce DCI.^20,21 Multiple end points were assessed, including CV, microvascular spasm, microthrombosis, CGRP and intercellular adhesion molecule 1 (ICAM-1) concentrations, degree of cerebral ischemia and apoptosis, and neurobehavioral outcomes.

METHODS

Experimental Animals

All experimental protocols were approved by the Institutional Animal Care and Use Committee of the research institute and were performed in accordance with the National Institutes of Health guidelines for the use of animals and ARRIVE guidelines for reporting animal research. A total of 107 adult male Sprague-Dawley rats (Taconic Biosciences Inc), weighing 300 to 375 g, were housed in temperature-controlled conditions with 12-hour light/dark cycle and free access to food and water. All procedures and assessments occurred in the same area and in the same order, so as to limit confounders.

SAH Model and Induction of CSDs

Endovascular perforation combined with artificially induced CSDs was performed as described.²¹ Briefly, rats were anesthetized with isoflurane and a sharpened 3-0 Prolene suture was used to perforate the left middle cerebral artery (MCA) to induce SAH. Endovascular puncture caused a sudden drop in CBF by 87% ± 11% (n = 5) for the first 5 minutes, which returned to baseline by 25 minutes (Figure 1A). At 24 hours after SAH, rats were reanesthetized and placed in a stereotaxic head frame. A 1.5-mm craniotomy was performed at 2 mm anterior to bregma and 2 mm left of the sagittal suture, for a 1 mol/L KCl-soaked cotton ball to artificially induce CSDs. Another 1.5-mm craniotomy was performed at 2 mm posterior to bregma and 2 mm left of the sagittal suture, for a laser Doppler flowmetry probe. After baseline recording, the cotton ball was applied to the cerebral cortex every 15 minutes for 1 hour (Figure 1B and 1C). Rats in the sham-operated group underwent all above procedures except for suture perforation to induce SAH. SAH grade was obtained according to the Sugawara grading system.²² At 48 hours after SAH, subarachnoid blood clots were mainly found around the circle of Willis (Figure 1D). The average SAH severity was 12.6 ± 2.8, with no difference in SAH severity between SAH-control and SAH-TNS groups (P = .6).

FIGURE 1.

Experimental designs. A, CBF response in the distal middle cerebral artery territory during SAH induction. Upon endovascular perforation, CBF rapidly declined and then recovered to baseline after approximately 25 minutes. B, SAH was induced by endovascular perforation on day 0. On day 1, 1 mol/L KCl was applied to artificially induce CSDs, while either TNS or sham stimulation were delivered. On day 2, rats were euthanized, and brain samples were collected for analysis. C, Diagram demonstrating CSD induction, CBF/CSD measurement, and TNS sites. D, Whole-brain photographs taken at 48 hours after SAH induction show thick hemorrhage layering the caudal and rostral portions of the cerebrum after successful endovascular puncture. CBF, cerebral blood flow; CSD, cortical spreading depolarization; ECoG, electrocorticography; KCl, potassium chloride; SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation.

Percutaneous TNS

A single bipolar electrode (concentric-bipolar 26G electromyography needle electrode, Natus Neurology Incorporated) was inserted percutaneously into the left infraorbital nerve and stimulated using an electrical stimulator (Isolated Pulse Stimulator Model 2100, A-M Systems).¹⁹ Rectangular biphasic pulses were delivered (50 Hz, 0.5-2.5 V) for a period of 20 seconds every 5 minutes over a 60-minute stimulation session that occurred concurrently with CSD generation (Figure 1B). Sham-operated and SAH-control animals had needle electrodes placed but did not receive electrical stimulation. CBF changes were measured by calculating the difference between baseline-normalized CBF (0 minute) and final CBF value (75 minutes).

Specimen Preparation and Measurements

Procedures for specimen preparation and quantification of vessel diameters, wall thickness, microthrombi, hypoxia inducible factor 1 subunit alpha (HIF-1α) and Tdt-mediated deoxyuridine triphosphate nick-end labeling (TUNEL)-positive neurons,²³ measurement of CGRP and ICAM-1 levels, and Garcia neuroscore²⁴ are detailed in Text, Supplemental Digital Content, http://links.lww.com/NEU/B417.

Experimental Groups and Exclusion Criteria

Rats were randomly divided into the sham-operated (n = 22), SAH-control (n = 50), and SAH-TNS groups (n = 30). One control group receiving neurectomy of trigeminal nerve branches (SAH-trigeminal neurectomy [Tx]) (n = 5) was added to confirm the specificity of TNS in releasing CGRP and preventing cerebral ischemia. Data from animals meeting any of the following exclusion criteria were removed from all statistical analyses per our protocol: (1) animals that did not survive through the entirety of the experimental period (n = 21 from SAH-control; n = 7 from SAH-TNS); and (2) absence of subarachnoid blood at 48 hour in SAH groups (n = 2 from SAH-control; n = 3 from SAH-TNS).

Randomization

To randomize animals' assignment to “SAH-control” vs “SAH-TNS” groups, we used random number generator. If the generated number is odd, then the animal is assigned to “SAH-TNS” group. Similarly, even number is used to assign animal to “SAH-control” group. All assessments were performed by individuals blinded to the experimental groups.

Data and Statistical Analysis

Data were expressed as means ± standard deviation and analyzed using GraphPad Prism software. All groups were analyzed for normality using the Shapiro-Wilk method. Effect sizes were quantified using Cohen's d for Student's t-test. Comparisons between 2 groups were made with Student's t-test, with a significance level of less than 0.05. Data that support the findings of this study are available upon reasonable request.

RESULTS

TNS Mitigates CSD-Induced Cerebral Ischemia

CSDs in SAH lead to impaired autoregulation and neurovascular decoupling.^25–27 Artificially generated CSDs increased CBF at 75 minutes in sham-operated brains; however, it resulted in decreased CBF in SAH brains; TNS maintained CBF (118% ± 15% vs 85% ± 10% vs 106% ± 11%; sham-operated [n = 7] vs SAH-control [n = 13] vs SAH-TNS [n = 10]; Figure 2A). Neurectomy of trigeminal nerve fibers (n = 5) abolished TNS′ ability to prevent CSD-induced cerebral ischemia.

FIGURE 2.

Impact of TNS on CBF, CSF CGRP, and cortical ICAM-1 concentration at 48 hours after induction of SAH. A, TNS preserved CBF after 60 minutes of successive CSDs. There was a marked decrease in CBF without TNS (SAH-control and SAH-Tx). B, SAH-control and SAH-Tx rats exhibited increases in CSF CGRP levels compared with sham-operated rats, whereas SAH-TNS rats exhibited a significantly greater increase in CSF CGRP levels. C, SAH-control rats had a significant increase in cortical ICAM-1 levels when compared with sham-operated rats. The TNS treatment significantly decreased ICAM-1 expression compared with the SAH-control rats. There was no difference between ICAM-1 levels in the sham-operated and SAH-TNS rats. *P < .05; **P < .01; ***P < .001; ****P < .0001. CBF, cerebral blood flow; CGRP, calcitonin gene-related peptide; CSD, cortical spreading depolarization; CSF, cerebrospinal fluid; ICAM-1, intercellular adhesion molecule 1; SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation; Tx, trigeminal neurectomy.

TNS Increases Cerebrospinal Fluid CGRP Concentration

SAH-control rats demonstrated a 98% increase in CGRP concentration as a response to severe vasoconstriction associated with SAH-induced CV (6.0 ± 3.2 vs 11.9 ± 2.9 pg/mL; sham-operated [n = 8] vs SAH-control [n = 9]). TNS increased CGRP levels by 89% compared with SAH-control rats (22.5 ± 3.6 pg/mL; SAH-TNS [n = 7]) (Figure 2B). Neurectomy abolished the effect of TNS to release CGRP (9.1 ± 4.2 pg/mL; SAH-Tx [n = 5]).

TNS Decreases Cortical ICAM-1 Expression

ICAM-1 is an early and specific marker for vasospasm.²⁸ Cortical ICAM-1 expression was significantly increased after SAH, and TNS treatment significantly decreased its expression (1010 ± 210 vs 2110 ± 560 vs 1290 ± 490 pg/mg protein; sham-operated [n = 5] vs SAH-control [n = 6] vs SAH-TNS [n = 6]; Figure 2C).

TNS Protects Against SAH-Induced CV

Significant large-vessel vasoconstriction was observed in SAH animals (Figure 3A). Cerebral vessels in SAH-control rats displayed significant decreases in vessel diameter (internal carotid artery [ICA]: 192 ± 28 µm vs 148 ± 49 µm; MCA: 224 ± 50 µm vs 144 ± 45 µm; anterior cerebral artery [ACA]: 189 ± 32 µm vs 132 ± 30 µm, sham-operated [n = 7] vs SAH-control [n = 13]), with reductions of 23%, 36%, and 30%, respectively. TNS increased the luminal diameters of ICA, MCA, and ACA by 50%, 33%, and 19%, respectively (ICA: 222 ± 38 µm; MCA: 192 ± 41 µm; ACA: 157 ± 36 µm; n = 8; Figure 3B-3E).

FIGURE 3.

Cerebral vasospasm was improved with TNS treatment at 48 hours after induction of SAH. A, Representative photomicrographs of H&E-stained cross-sections of the BA, left ICA, left MCA, and left ACA on day 2 are shown. Scale bar = 200 µm. B, Vessel diameter measurements of BA. C, Vessel diameter measurements of ICA. D, Vessel diameter measurements of MCA. E, Vessel diameter measurements of ACA. SAH-control rats demonstrated mild to moderate vasoconstriction, which was alleviated with TNS treatment. *P < .05; **P < .01; ***P < .001. ACA, anterior cerebral artery; BA, basilar artery; ICA, internal carotid artery; MCA, middle cerebral artery; SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation.

TNS Protects Against Vasoconstriction of Cortical Arterioles

Because of large variations in the pial arteriole diameter (98 ± 32 µm vs 46 ± 28 µm vs 56 ± 23 µm, sham-operated vs SAH-control vs SAH-TNS), we measured the wall thickness of arterioles (Figure 4A).²⁹ Pial arteriolar wall thickness at 48 hours after SAH was significantly greater than that of sham-operated rats (8.1 ± 1.6 µm vs 12.0 ± 3.6 µm, sham [n = 7] vs SAH-control [n = 13]). TNS treatment significantly decreased arteriolar wall thickness (9.2 ± 1.8 µm; n = 8) (Figure 4B).

FIGURE 4.

Pial arteriolar wall thickness and microthrombi at 48 hours after induction of SAH. A, H&E-stained photomicrographs of cortical pial arterioles demonstrate corrugation of the internal elastic lamina and thickening of the tunica media and tunica intima with associated luminal narrowing after SAH (magnified in solid black box). Treatment with TNS partially relieves this effect. Numerous eosinophilic aggregates indicative of microthrombi are identified in the parenchyma of SAH-control rats, but are significantly reduced in SAH-TNS rats (magnified in dashed red box and indicated by red arrows). B, Measurements of cortical pial arteriolar walls demonstrate improved wall thickness after TNS treatment. C, Quantification of microthrombi reveal substantially fewer microthrombi in SAH-TNS rats, compared with SAH-control rats. *P < .05; **P < .01; ***P < .001; ****P < .0001. ROI, region of interest; SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation.

TNS Reduces Cerebral Microthrombi Formation

Microthrombi counts were significantly greater in SAH-control rats compared with sham-operated rats (10.6 ± 6.3 vs 51.7 ± 20 microthrombi per region of interest, sham-operated [n = 7] vs SAH-control [n = 13]). TNS treatment dramatically reduced microthrombi counts by 53% (24.5 ± 4.5 microthrombi per region of interest; n = 8; Figure 4A and 4C).

TNS Reduces Neuronal Ischemia and Apoptosis

After SAH, neurons subjected to ischemia display upregulation of HIF-1α, which is a measure of hypoxic stress and whose early-stage activation may be detrimental.³⁰ SAH-control rats demonstrated significant HIF-1α immunopositivity at 48 hours after induction of SAH (0.4 ± 0.5 vs 45.5 ± 23 neurons per high powered field (HPF), sham-operated [n = 5] vs SAH-control [n = 6]). SAH-TNS animals displayed substantially reduced HIF-1α immunopositivity (10.2 ± 8.5 neurons per HPF; n = 5; Figure 5A). SAH-control animals displayed an exceedingly high level of TUNEL-positive cells compared with sham-operated animals (2.2 ± 0.8 vs 62.5 ± 16 neurons per HPF, sham-operated [n = 5] vs SAH-control [n = 6]). TNS reduced the number of TUNEL-positive neurons by 69% compared with SAH-control animals (19.6 ± 11.1 neurons per HPF; n = 5; Figure 5B).

FIGURE 5.

HIF-1α and TUNEL-positive apoptotic cells in CA1 region of the left hippocampus at 48 hours after induction of SAH. A, The changes of HIF-1α, NeuN, and DAPI expression after 48 hours of SAH. Green, red, and blue IF signals represented HIF-1α, NeuN, and DAPI, respectively. The colocalization of HIF-1α/NeuN/DAPI was quantified. SAH-control rats demonstrated a high degree of HIF-1α positivity within the CA1 region of the left hippocampus after SAH. TNS treatment greatly decreased the number of HIF-1α immunopositivity (arrow). B, TUNEL-labeled and DAPI-stained nuclei of the left hippocampal CA1 region. Green and blue IF signal represented TUNEL and DAPI, respectively. The colocalization of DAPI/TUNEL in combination with the morphological features of the labeled nuclei indicating an apoptotic phenotype was quantified. Significant TUNEL-labeled nuclei was observed in hippocampal CA1 pyramidal neurons of SAH-Control animals, consistent with a high degree of neuronal apoptosis after SAH-induced CV. TNS treatment significantly reduced apoptotic cells in hippocampal neurons. *P < .05; **P < .01; ***P < .001; ****P < .0001. CV, cerebral vasospasm; DAPI, 4',6-diamidino-2-phenylindole; IF, immunofluorescence; NeuN, neuronal nuclear protein; SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation; TUNEL, tdt-mediated deoxyuridine triphosphate nick-end labeling.

TNS Ameliorates SAH-Induced Neurological Deficits

Sham-operated rats showed a stable performance at 24 and 48 hours, whereas the SAH-control rats showed a significant decrease in sensorimotor function (Figure 6). The neurological score was significantly improved in the SAH-TNS group compared with the SAH-control group at 48 hours (8.2 ± 0.9 vs 12.9 ± 1.6; SAH-control [n = 13] vs SAH-TNS [n = 8]).

FIGURE 6.

Neurological outcomes. Neurobehavioral testing was performed using the Garcia Neuroscore on all rats immediately before SAH induction (0 hour), cortical spreading depolarization induction with or without TNS (24 hours), and euthanasia (48 hours). At 24 hours, both SAH-control and SAH-TNS rats showed equivalent reductions in neurological scores. At 48 hours, TNS-treated rats displayed a greater improvement in Garcia score than SAH-control rats. *P < .05 vs sham; #P < .05 vs SAH-control. SAH, subarachnoid hemorrhage; TNS, trigeminal nerve stimulation.

DISCUSSION

DCI, a multifactorial disorder with a combination of large artery vasospasm and microcirculatory failure, is one of the most debilitating complications of SAH.¹ With the failure of clinical studies that have targeted CV to date, attention has recently been redirected to other potential causes of DCI, such as microvascular dysfunction, microthrombosis, CSDs, and neuroinflammation.^12,23 In particular, microcirculatory dysfunction and impaired tissue-level perfusion are emerging as important players in this complex web of pathological changes after SAH, ultimately leading to spreading ischemia when spontaneous CSDs occur.^{7,12,13,31–33} To evaluate the effect of TNS on both cerebral macrocirculation and microcirculation, we replicated the multifactorial nature of DCI using a rat endovascular puncture model of SAH combined with artificially induced CSDs, which more closely emulates the post-SAH syndrome seen in the clinical setting.^21,34–36 We demonstrate that TNS not only resolves large-vessel vasospasm, but also improves microcirculatory perfusion, ultimately resulting in better neurological outcomes. This broad capability of TNS to act across multiple levels of cerebral circulation, leading to improvements in tissue-level perfusion and preventing ischemic injury leading to DCI, has never been described.

There is growing evidence demonstrating that CV alone is not the only pathological process that results in worse clinical outcomes; however, relieving CV and improving CBF remains a key focus of current interventions for DCI.³ The results presented herein demonstrate that TNS clearly improves large artery vasospasm (ICA/MCA/ACA) in the setting of SAH by eliciting robust vasodilation in cerebral vessels. Furthermore, TNS mitigates CSD-induced cerebral ischemia in SAH brains. This finding is consistent with results of previous studies that TNS results in cerebral vasodilation and CBF elevation in SAH brains.^17–19 Although the role of TNS against CV is becoming clearer, the underlying mechanisms by which TNS enhances cerebral circulation after SAH remain to be fully elucidated. The effect is likely driven by the release of CGRP, a potent vasodilatory neuropeptide.^37–39 Despite preclinical studies suggesting positive effects of exogenous CGRP administration after SAH, clinical studies have largely failed to show benefit.^40–42 This may be because of different methods of CGRP delivery between preclinical and clinical protocols; preclinical studies have used intranasal,⁴³ intracisternal,^44,45 or intrathecal administration,^46,47 whereas clinical studies have relied on intravenous administration. CGRP has a short half-life (approximately 7-10 minutes) when it is introduced into systemic circulation,⁴⁸ whereas it has been shown to have much longer-lasting effects (approximately 4-6 hours) when released into cerebrospinal fluid (CSF).^46–48 Here, we show that TNS may be particularly beneficial in acute vasoconstrictive conditions such as CV, because of its focal release of CGRP in the brain. This study demonstrates that the neuroprotective effect of 1-hour TNS delivered at 24 hours after SAH lasts at least 48 hours. This may be because TNS increases CSF CGRP concentration, leading to much longer-lasting effect.¹⁹

In addition to CV, major morphological changes in the pial microcirculation occur in the early stage of SAH.⁴⁹ Acute vasoconstriction occurs predominantly in small arterioles, alongside a progressive increase in the number of adherent leukocytes in small vessels, contributing to reductions in CBF to cortical brain tissue after SAH. Damaged endothelial cells exposed to reactive oxidative species form blood breakdown products and inflammation, and release prothrombotic factors, such as von Willebrand factor and fibrinogen, resulting in the accumulation of microthrombi within pial arterioles.¹¹ As these platelet aggregates propagate, distal microvasculature constricts, and ischemia ensues, strongly correlating with neuronal apoptosis. Although microcirculatory dysregulation clearly impairs cerebral perfusion as demonstrated in both preclinical⁷ and clinical^8–11 studies, these phenomena are difficult to therapeutically target.⁵⁰ In this study, we have shown that TNS reduced microvasospasm of cortical arterioles, microthrombosis formation, and cortical ICAM-1 levels by 23%, 53%, and 39%, respectively. It is likely that these effects are mediated by the release of CGRP, which has shown a vasodilatory effect on pial arterioles.^51–53 Our results are consistent with previous findings that intravascular CGRP dilates small parenchymal vessels,^54,55 and the increase in CGRP is associated with a decrease in ICAM-1 expression.⁵⁶ However, it may be also mediated by other vasoactive agents because the free nerve endings of the trigeminal system are known to release a host of vasoactive agents,¹⁶ such as nitric oxide,^57,58 pituitary adenylate cyclase–activating polypeptide,^59,60 and epoxyeicosatrienoic acids,^61,62 which interact directly with endothelial cells to reduce inflammation, reactive oxidative species–mediated damage, and alter the thrombogenicity profile of vascular endothelium. It is our opinion that TNS may much more effectively enhance cerebral microcirculation compared with the administration of a single vasoactive peptide, such as CGRP, by generating a myriad of vasoactive agents targeting different pathological mechanisms of DCI.

By addressing large artery vasospasm, as well as microcirculatory dysfunction, TNS ultimately led to decreases in ischemia-induced hypoxic neuronal injury, as demonstrated by decreased HIF-1α immunopositivity and neuronal apoptosis resulting in improved functional outcomes. Although HIF-1α is fundamentally a measure of hypoxic damage, its production is upregulated by the presence of ischemia, whereupon it regulates multiple ischemia-affected processes, including glucose transport and oxidative stress.³⁰ As such, its upregulation is intrinsically linked to the degree of neuronal damage and functional impairment. Based on both our results and those seen in injured brains, it is likely that the increased CGRP levels are performing secondary activations of different pathways beneficial for the postinjured brain, such as MAPK,⁶³ AKT/MTOR,^64,65 WNT/β-catenin,⁶⁶ and FoxO3a.⁶⁴ These pathways have been shown to play an important role in cognitive function after brain injuries. Delayed ischemic neurological deficits (DIND) are a devastating consequence of SAH, which affects about 30% of surviving patients.⁶⁷ Impaired cerebral microcirculation results in capillary dysfunction leading to microvasospasm of pial arterioles and a subsequent reduction of parenchymal microperfusion, which significantly contribute to DIND, independent of CV. Performing secondary activation of neuroprotective pathways may be another important approach to prevent DIND after SAH.

Limitations

There are several limitations to this study. First, our experiments were conducted with a small animal model. There are limitations to each animal model of SAH, which contribute to the difficulty in directly translating preclinical findings to effective therapies.⁶⁸ Although we tested TNS in a clinically relevant animal model by the addition of CSDs, it warrants further testing in large animal models that may better reproduce the pathophysiology of DCI in patients with SAH. Second, small animals develop delayed neurological deficits 3 to 6 days after SAH.⁶⁹ Most findings discussed herein represent outcomes at day 2, indicating the necessity of future work investigating neurological outcomes at longer time points. Third, we have only tested the effect of a single dose of TNS at a single time point. Future work should focus on determining the optimal dose of TNS and its therapeutic time window on the progression of DCI. Finally, whether or not CGRP is the key player in TNS-induced protection from DCI needs to be further investigated with pharmacological blockade.

CONCLUSION

Our study demonstrates that TNS activates a powerful cerebrovascular response to vasospasm, which improves both macrocirculatory and microcirculatory perfusion after SAH, leading to a reduction in cerebral ischemia and better functional outcomes. In addition to increasing CGRP concentration in the brain, TNS may upregulate a myriad of other vasoactive agents and promote activation of other neuroprotective pathways for preventing DCI.

Funding

This work is supported by National Institute of Neurological Disorders and Stroke of the National Institutes of Health under award number R21NS114763 and the US Army Medical Research and Materiel Command (USAMRMC) under award #W81XWH-18-1-0773.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.

Supplemental Digital Content

Supplemental Digital Content. Text. Detailed experimental procedures. Methods for specimen preparation, as well as quantification of vessel diameters, wall thickness, microthrombi, HIF-1α and TUNEL-positive neurons, measurement of CGRP and ICAM-1 levels, and behavioral outcomes.