A concussion is defined as a traumatic brain injury induced by biomechanical forces (1) and is estimated to account for more than 3000 hospital emergency department visits each day in the United States (2). Common signs and symptoms after concussion include somatic, cognitive, emotional, physical, behavioral, or sleep disturbances (1). Typical concussion-related symptom resolution occurs within 14 d for 90% of young and adult sport-related injuries (3,4), with symptom resolution serving as one benchmark to a return to physical activity (1). Although many individuals may experience rapid improvements in symptom recovery and return to physical activity after concussion, the neurophysiological recovery is likely incomplete. The emergence of advanced neuroimaging studies suggests that concussion-related symptomology, neurocognitive performance, and balance may not necessarily reflect neurophysiological recovery and that healing may occur up to 14 d after injury or longer (1,5).
A neurometabolic cascade of events occurs immediately after concussion and has been suggested to persist between 10 and 30 d after injury (5–7). After a concussive insult, the brain enters a hypermetabolic state with increased glucose and oxygen demand (8) with a concurrent decrease in cerebral blood flow (5). Cerebral blood flow has been reported to be reduced up to 50% of preinjury values throughout the recovery from a concussion. Reduced cerebral blood flow restricts glucose and oxygen delivery and subsequent adenosine triphosphate production. Adenosine triphosphate production is necessary to correct the ionic imbalance that occurs after a concussive insult (5). Various interventions have been suggested to mitigate the deleterious pathophysiological changes after concussion and may facilitate a “typical” recovery (i.e., ≤14 d) and complete recovery improving patient outcomes (9). One such modality that may produce these therapeutic effects is cryotherapy.
Cryotherapy has been studied extensively for its use throughout the musculoskeletal system and has been demonstrated to acutely decrease superficial and deep tissue temperature, oxygen, and energy demands (10–12). Systemic cooling has been reported to reduce metabolic demand, cerebral blood flow, and improved function recovery in moderate and severe traumatic brain injury in animal and human models (13–15). More specifically, reduced cerebral tissue temperature has been demonstrated to reduce energy and metabolic requirements, an effect that may be advantageous after concussion (16).
Although there is evidence for systemic cooling modalities, little is known regarding the effects of superficial cranial cryotherapy on cerebral hemodynamics and metabolism after concussion. Manufacturers of cryotherapy equipment have purported that cranial hypothermia using a gel-filled cap may be therapeutic after concussion (17). These claims have only been assessed empirically in one study comparing healthy and concussed participants using functional magnetic resonance imaging (18); however, the physiological responses to superficial cranial cooling have not been investigated. A critical step before implementing treatment interventions in a pathological group is to establish its safety and physiological response. Poor understanding of an intervention could have deleterious effects on patient outcomes, or may not provide any physiological benefit, thus making the intervention ineffective. To promote safe, evidence-based medicine when treating concussions, initial investigations among healthy participants are warranted to establish their safety and efficacy first.
Therefore, the purpose of our study was to determine the effects of superficial cranial cryotherapy on cerebral hemodynamics and neurocognitive performance in healthy participants. We hypothesized that individuals who received superficial cryotherapy would have no change in cortical hemodynamics or neurocognitive performance due to highly regulated cerebrovascular control. As an early study examining superficial cryotherapy on cortical hemodynamics, we wanted to determine the treatment’s effects on a healthy sample.
The current study was a single-blinded parallel randomized control trial conducted in a university sports medicine laboratory. A convenience sample of 34 healthy participants were initially assessed for eligibility and enrolled in the study from a large public university. All 34 participants were randomly allocated to groups (n = 15 cryotherapy, n = 19 control), and all participants were included for analyses. Participants were included if they were between 18 and 25 yr. Participants were excluded if they self-reported any contraindications to cryotherapy (e.g., Raynaud’s phenomenon or cold urticaria), any diagnosed learning difficulty or disability, a history of migraines, a history of a concussion or traumatic brain injury in the 6 months before study participation, current use of psychotropic medications, or diagnosed with any neurological, psychiatric, metabolic, hematological, or cardiopulmonary disease. Participants who reported exercise, caffeine or alcohol consumption, or a headache within 12 h before study participation were deferred to a later testing time. There were no changes to the study design or outcome measures after recruitment of the first participant. This study was approved by the institutional review board and was registered in the US National Institutes of Health clinical trials registry (NCT03185507).
Functional Near-Infrared Spectroscopy
Cortical hemodynamics were measured using an Artinis PortaLite functional near-infrared spectroscopy (fNIRS) unit and OxySoft software (Artinis Medical Systems, B.V., Elst, the Netherlands) providing outcome measures for oxygenated hemoglobin (HbO) and deoxygenated hemoglobin (HbR). The use of fNIRS to measure cortical hemodynamics has been demonstrated to be reliable and ideal for noninvasive measures of oxygen saturation in the superficial cortex (19,20).
Superficial Skin Temperature
Skin surface temperature was assessed using the THERMES USB thermocouple data acquisition system and three PT-6 thermocouple probes (Physitemp Instruments Inc., Clifton, NJ).
Superficial cryotherapy was administered using the Catalyst CryoHelmet™ (All-Star Sporting Goods®, Shirley, MA; Figs. 1A, B). The CryoHelmet™ is a commercially available cranial cooling device and was selected for the intervention owing to its simplicity, consistent application, and open availability to the public. The CryoHelmet™ was stored in a freezer at −20°C.
Neurocognitive performance was assessed using a computerized Stroop Test (ST; CNS Vital Signs, LLC., Version 4, Morrisville, NC) installed on a desktop computer. The ST has been demonstrated to be a reliable and valid clinical measure of executive function and selective attention, which provides focal neuronal activity in the prefrontal cortex (21,22). The ST is a three-part test that provides outcome measures for simple reaction time, complex reaction time correct, Stroop reaction time, Stroop commission errors, and overall reaction time by taking the average time between word presentation and correct motor response (i.e., pressing a computer space bar) for each corresponding trial (22). Simple reaction time is calculated from trials where a neutral colored word appears, complex reaction time is calculated from trials where words of a color are written in their corresponding color (e.g., the word “red” is written in red color), Stroop reaction time is the inverse of the complex reaction time (e.g., responding when the word “red” is written in any color BUT red), overall reaction time is the participant’s average time to respond during the complex reaction time and Stroop reaction time combined, and Stroop commission errors are the total number of errors throughout the Stroop reaction time (22). The ST was administered two times during the study protocol, once before the intervention and again postintervention.
After obtaining consent, participants provided their medical history and completed the revised Head Injury Scale (HIS-r) (23), Hospital Anxiety and Depression Scale (HADS) (24), Perceived Stress and Reactivity Scale (PSRS) (25), and a visual analog scale (VAS) rating their current physical overall pain (26). After determining if a participant met the inclusion criteria, the participant remained seated quietly at the testing station for 10 min before fNIRS data collection. The testing station consisted of a desktop computer in a quiet, secluded laboratory room with a dividing partition in place to limit distractions and extraneous cognitive stimulation. All participants were assessed in the morning to mitigate influences of daily physical and cognitive activity on cerebral perfusion or the potential of headaches from caffeine withdrawal later in the day.
Participants were equipped with fNIRS by affixing an optode over the left superficial prefrontal cortex location (Fig. 1C). The prefrontal cortex location was determined by measuring 2.54 cm superior and lateral to the glabella anatomical landmark. Black foam prewrap (Cramer Products Inc., Gardner, KS) was used to hold the fNIRS optode to the head and to block ambient light. Three thermocouples sampling data at 2 Hz were used and placed on the laboratory floor for room temperature, over the left carotid artery, and left temporal artery affixed with steri-strips (3 M, St. Paul, MN) and adhesive spray (Cramer Products Inc.; Fig. 1C). Thermocouple location for the left carotid artery was lateral to the thyroid cartilage at the point of the strongest palpable pulse. The thermocouple for the temporal artery was positioned adjacent to the pterion on the left temple. The thermocouple unit was active for at least 20 min before data collection per manufacturer’s recommendations to ensure measurement accuracy. After the rest period, the fNIRS optode and thermocouples were activated to collect data and remained active for the duration of the study visit. The fNIRS optode collected at a sampling rate of 10 Hz and emitted continuous infrared light at 762 and 853 nm from three light sources and one detector at fixed distances of 30 (source 1), 35 (source 2), and 40 (source 3) mm from the detector. These source-to-detector distances correlate with penetrating depths of approximately half the respective distance (27). To calculate the scattering coefficient for fNIRS, the following formula was used:
where the constant h (nm−1) was set to 4.5 • 10−4 and k (mm−1) was set to 1.63 based on previous literature (28). The fNIRS differential pathlength factor was calculated using a validated formula by Duncan et al (29):
The same clinician performed all baseline and posttreatment assessments (L.B.L.) and was blinded to treatment group allocation. Superficial temperature and cortical hemodynamics were recorded during the duration of the study (Fig. 2) and separated into seven epochs: 1) 5 min of cognitive rest, 2) during the ST (approximately 6 min), 3) cognitive recovery for 5 min after the ST, 4) during the 20-min intervention period, 5) 5 min of cognitive rest after intervention, 6) during the postintervention ST, and 7) 5 min after the postintervention ST. When the ST was not being administered, the computer monitor was covered with a screen cover with a 5-cm diameter black circle in the center. Participants were instructed to look at the black circle throughout cognitive rest and intervention time points to reduce additional cognitive activity that may cause hemodynamic artifact from ocular function.
Treatment randomization was performed a priori using a random number generator (J.J.F.) and was stratified on sex. Treatment allocation was concealed in a sealed, opaque envelope. Before the intervention, the clinician conducting the assessments (L.B.L.) left the room. For all participants, a second investigator (J.J.F. or N.K.E.) carried the cryotherapy device in an insulated case from the freezer to the testing station, opened the sealed opaque allocation envelopes, and administered the assigned treatment. Participants allocated to the intervention (cryotherapy) group received superficial cryotherapy applied to the participant’s head and neck for 20 min (Figs. 1A, B). Participants allocated to the control group did not receive any treatment and were asked to remain seated quietly for 20 min. The intervention time frame was selected based on previous literature demonstrating that 20 min was effective at decreasing skin temperature and metabolic activity in deep tissue (11). After the 20-min intervention, the cryotherapy device was removed. The assessing clinician (L.B.L.) returned and continued recording superficial temperature and cortical hemodynamics measures for 5 min, whereas the participants remained in a static quiet-sitting position for a postintervention cognitive rest. After the postintervention cognitive rest, a second ST was administered and followed by a final 5 min of cognitive rest. After the 5 min of cognitive rest after intervention, participants completed a final VAS and HIS-r. After the final assessment time point, the participants were disconnected from all recording devices and dismissed.
Data from the fNIRS unit were exported from Artinis Oxysoft software to MATLAB (Mathworks®, Natick, MA) for processing using the Hemodynamic Evoked Responses 2 v.1.5.2 (HomER2) plugin for MATLAB (30). Using HomER2, subjects were sorted into their allocation, and each subject’s start and end times and stimulation for time points were marked in the program for analysis. The raw fNIRS signals were processed into optical density and then relative concentrations of HbO and HbR. All data were processed to remove motion and physiological artifact and filtered with a high- (0.010 Hz) and low-bandpass filter (0.5 Hz). The HbO and HbR were visually examined for an inverse relationship, indicating that the signal was processing cortical responses and not motion artifact. All fNIRS cognitive rest and intervention time point data had 10% of fNIRS data at the beginning and end of each time epoch excluded from analysis to reduce motion artifact as a result of verbal instructions or transitions through the study protocol.
A priori power analysis indicated 16 participants were needed to demonstrate large treatment effects based on HbR (SD = 0.38 μmol·L−1, α = 0.05, and β = 0.80) from a previous study using a similar technique evaluating cortical hemodynamics (31).
Independent t tests were used to assess demographic and baseline values for all variables between groups. Data were assessed for outliers, multivariate normality, and homogeneity of variance before analysis. A 2 (group) × 7 (time) repeated-measures ANOVA for each fNIRS and thermal dependent variables (HbO, HbR, carotid artery temperature, temporal artery temperature) to compare before, during, and after intervention outputs between and within groups. A 2 (group) × 2 (time) repeated-measures ANOVA was conducted for the neurocognitive outcomes (simple reaction time, complex reaction time, Stroop reaction time correct, Stroop commission errors, and overall reaction time). If statistically significant, post hoc paired t tests were calculated for each dependent variable. Bonferroni adjustments were made to the hemodynamic and thermal outcomes because of numerous pairwise comparisons from the post hoc 2 × 7 repeated-measures ANOVA. Overall reaction time from the ST was assessed within both study groups for any clinically meaningful change as indicated by mean ∆ score from preintervention to postintervention being >5% of the preintervention average. All statistical analyses were conducted using SPSS software Version 184.108.40.206 (SPSS Inc., Chicago, IL), and α was set to 0.05 for all analyses a priori.
No significant differences were observed between the control group (n = 19; 11 male, 8 female) and the cryotherapy group (n = 15; 6 male, 9 female) for any demographic, mood (PSRS and HADS), symptom (HIS-r), or global pain (VAS) variables (P > 0.05; Table 1). No differences were found between groups over time regarding the VAS and HIS-r (P > 0.05).
Room temperature over time and in both groups did not change during the testing protocol (24.4°C ± 1.28°C, P > 0.05). Carotid temperature displayed a significant group–time interaction (F(1,35) = 17.8, P < 0.001, η2p = 0.53; Figure 3). Post hoc analyses revealed a significant carotid artery temperature decrease from baseline to intervention time points in the cryotherapy group by 5.4°C ± 5.7°C (P = 0.002), whereas the control group significantly increased by 0.3°C ± 0.4°C (P = 0.08). Temporal temperature had a significant group–time interaction (F(2,61) = 3.40, P < 0.001, η2p = 0.53), with post hoc t tests demonstrating the cryotherapy group significantly increased by 0.22°C ± 0.29°C (P < 0.001) and the control group increased by 0.35°C ± 0.2°C (P = 0.01) from baseline to intervention time points.
Stroop commission errors were significantly altered before to after intervention between groups (F(1,32) = 4.32; P = 0.05, η2p = 0.12), with the cryotherapy group committing statistically fewer errors compared with the control group (0.7 ± 0.26 vs 1.2 ± 0.23; Table 2). No significant differences were observed between groups for the simple reaction time, complex reaction time correct, or Stroop reaction time (P > 0.05). No clinically meaningful changes in overall reaction time were observed in either the cryotherapy (∆ score, 16.07; 5% threshold, 30.37) or control (∆ score, 3.53; 5% threshold, 28.87) groups.
All hemodynamic comparisons over time are presented in Figure 4. HbO displayed a significant group effect at approximately 40-mm cortical depth (F(1,32) = 4.21; P = 0.05, η2p = 0.12), significant time effect at approximately 35-mm cortical depth (F(1,32) = 7.94; P = 0.005, η2p = 0.20), and no significant differences at approximately 30-mm cortical depth (P ≥ 0.17). Post hoc t tests, however, demonstrated no significant differences in HbO among any of the light sources between the cryotherapy and control groups after Bonferroni corrections, indicating that superficial cryotherapy did not significantly alter HbO throughout the duration of the study.
HbR did not display any significant main or interaction effects at 40-, 35-, or 30-mm approximate cortical depths (P ≥ 0.18), indicating that superficial cryotherapy did not significantly alter HbR at any time point between or within groups.
Cryotherapy is commonly used for musculoskeletal injuries and has been shown to decrease metabolic demand during the inflammatory process (10–12). Clinicians have empirically suggested that cryotherapy for concussion may have similar advantages, yet no physiological effects have been documented. To our knowledge, our study is the first to examine the effects of superficial cryotherapy on cortical hemodynamics and neurocognitive function. Our goals were to examine the safety of future clinical trials involving concussed patients when applying cold therapy to the head, and to establish a protocol for evaluating discomfort, concussion symptoms, neurocognitive performance, and hemodynamic changes associated with the cryotherapy application.
Superficial cryotherapy reduced temperature over the carotid artery as observed in the differences between groups (Fig. 3) during and shortly after application, with gradual rewarming occurring. The reduction is tissue temperature after superficial cryotherapy is to be expected. However, we surprisingly observed a significant, though minimal, 0.22°C increase in temporal artery temperature between prerest and the intervention after cryotherapy with a plateau in tissue temperature beyond the intervention time point. This finding is likely the result of anatomical differences in head and neck anthropometrics, which limited the direct surface contact of the superficial cryotherapy over the temporal artery thermocouple site for each participant. The differences in temperature changes between the superficial carotid artery and temporal artery may also indirectly demonstrate that minimal thermal changes are possible on the cranium’s skin in healthy individuals, and may partially explain why hemodynamic differences in the brain were not different.
Our results indicate that cryotherapy did not affect neurocognitive performance based on the outcome measures of the ST. However, the cryotherapy group committed fewer errors on the second administration of the ST and fewer than the control group who committed on average 0.5 more errors. Though statistically significant, this finding has limited clinical meaning, as it is impossible to commit half of an error. Our results align with a recent study by Jackson et al. (32) examining neurocognitive performance after superficial cryotherapy, which demonstrated that neurocognitive performance is not altered after superficial cryotherapy. The lack of neurocognitive performance differences may be attributed to the ST itself, as it is only one of many psychometrically valid neurocognitive tests used clinically (22). The ST was selected for study use, however, as it 1) causes neuronal activation in the prefrontal cortex where the fNIRS device was placed, 2) is designed to test executive function (i.e., choosing to execute or inhibit responses), and 3) is frequently used in advanced neuroimaging studies (33).
The lack of cortical hemodynamic change after superficial cryotherapy may be due to the brain and surrounding tissue being well suited for preventing thermal change due to the external environment (34,35). Although previous research has demonstrated that superficial cryotherapy reduces metabolic activity in the extremities (10–12), it is unlikely that this is possible deeper in neuronal tissue after short-term superficial cryotherapy application and where thermal changes are highly regulated. Assessing solely neuronal tissues metabolic activity is not possible in vivo in humans, and thus, surrogates, such as HbO and HbR, are necessary. Analyzing HbO and HbR allows us to measure tissue perfusion and describe the relative oxygenated tissue saturation of the examination site, which is driven by metabolic demand, an impaired process after concussion (5–7).
To date, a dearth of literature has examined the effects of cryotherapy using fNIRS in a healthy population, which limits our ability to extrapolate our results to a clinical sample. Our results demonstrated that cryotherapy did not alter cortical hemodynamics as measured by HbO and HbR. No significant changes in HbO are a positive finding in the healthy cohort, as it indicates that superficial cryotherapy does not reduce blood flow or oxygen content to the brain. Although normally impairing oxygen concentrations in the brain is deleterious, decreasing oxygen utilization and ultimately metabolic activity after concussion may be beneficial. Concussions cause the brain to become hypermetabolic marked by increased glucose and oxygen demand (8); however, it may be advantageous to reduce metabolic activity acutely after the injury. Theoretically, if the metabolic activity could be safely throttled, this could make metabolic activity and cerebral blood flow match demands and ultimately reduce the neurometabolic cascade (5).
Our findings also indicated that HbR was not significantly influenced after superficial cryotherapy, but did show a trend of increased HbR during the intervention and postintervention ST. This finding may be due to a hemodynamic change occurring, or more likely as a result to the pressure applied to the internal jugular vein and carotid arteries via the cryotherapy device. The cryotherapy device used in this study provides light jugular vein compression (Fig. 2A), which may have altered cerebral blood flow by creating back-flow pressure that increased HbR concentration. Jugular vein compression devices have been purported to mitigate microstructural changes after repetitive head impacts and moderate traumatic brain injury (36,37). However, the authors of these studies did not investigate the blood saturation during jugular vein compression. The purpose of the venous system is to return relatively deoxygenated blood back to the heart then lungs for oxygenation. Creating back flow pressure via jugular vein compression may explain why a trend of increased HbR was observed. Regardless, an increase in HbR likely does not serve a physiological benefit, with the exception of potentially increasing blood volume and creating a “cushion” for the brain to reduce the risk for further damage (38,39). This rationale, however, is still theoretical, and further empirical research is necessary to support these claims.
The current study is not without its limitations. The fNIRS device used was a single-channel device examining a small focal point of the brain with a maximal examining depth of 2 cm on a capillary level, limiting the generalizability of the cryotherapy effect to the whole brain. Because of the focal examination of this device, it is also possible that the investigated brain region has different neurometabolic activity than an area a few millimeters away. Delimitations for our study consisted of the short, 20-min treatment time and use of a noncirculatory cryotherapy device. Although noncirculating cryotherapy allows for potential thermal gradients over time, we are confident that thermal changes did occur as evident through the thermocouple data. Lastly, the recruitment and use of a nonpathological cohort (e.g., concussed population) poses limited generalizability to these findings as well but was necessary to determine if physiological change occurs as a result of the application of cryotherapy before a clinical sample.
The purpose of this study was to explore whether cryotherapy had any effects on cerebral hemodynamics and neurocognitive performance in a healthy population. Our results indicate that 20-min of superficial cryotherapy does not alter cortical hemodynamics or neurocognitive performance when applied in a healthy population. Future studies should explore advanced neuroimaging to assess for global changes of the brain after superficial cranial cryotherapy in healthy and concussed participants to determine thermal and hemodynamic effects, and clinical utility of superficial cryotherapy.
The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. Results of the present study do not constitute endorsement by the American College of Sports Medicine. Neither the Department of the Navy nor any other component of the Department of Defense has approved, endorsed, or authorized this article.
All authors have no conflicts of interest or sources of funding to declare.
All participants provided informed consent, and this study was approved by the institutional review board for health sciences research at the University of Virginia in compliance with all applicable Federal regulations governing the protection of human subjects.
1. McCrory P, Meeuwisse W, Dvorak J, et al. Consensus statement on concussion in sport—the 5th international conference on concussion in sport held in Berlin, October 2016. Br J Sports Med
2. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil
3. McCrea M, Guskiewicz KM, Marshall SW, et al. Acute effects and recovery time following concussion in collegiate football players: the NCAA Concussion Study. JAMA
4. CARE Consortium Investigators, Garcia G-GP, Broglio SP, Lavieri MS, McCrea M, McAllister T. Quantifying the value of multidimensional assessment models for acute concussion: an analysis of data from the NCAA-DoD Care Consortium [Internet]. Sports Med
. 2018 [cited 2018 Mar 2]. Available from: http://link.springer.com/10.1007/s40279-018-0880-x
5. Giza CC, Hovda DA. The new neurometabolic cascade of concussion. Neurosurgery
. 2014;75(Suppl 4):S24–33.
6. Meier TB, Bellgowan PSF, Singh R, Kuplicki R, Polanski DW, Mayer AR. Recovery of cerebral blood flow following sports-related concussion. JAMA Neurol
7. Maugans TA, Farley C, Altaye M, Leach J, Cecil KM. Pediatric sports-related concussion produces cerebral blood flow alterations. Pediatrics
8. Yoshino A, Hovda DA, Kawamata T, Katayama Y, Becker DP. Dynamic changes in local cerebral glucose utilization following cerebral conclusion in rats: evidence of a hyper- and subsequent hypometabolic state. Brain Res
9. Wang H, Wang B, Jackson K, et al. A novel head-neck cooling device for concussion injury in contact sports [Internet]. Transl Neurosci
. 2015;6(1). [cited 2017 Feb 8]. Available from: http://www.degruyter.com/view/j/tnsci.2015.6.issue-1/tnsci-2015-0004/tnsci-2015-0004.xml
10. Dykstra JH, Hill HM, Miller MG, Cheatham CC, Michael TJ, Baker RJ. Comparisons of cubed ice, crushed ice, and wetted ice on intramuscular and surface temperature changes. J Athl Train
11. Ho SS, Illgen RL, Meyer RW, Torok PJ, Cooper MD, Reider B. Comparison of various icing times in decreasing bone metabolism and blood flow in the knee. Am J Sports Med
12. Minett GM, Duffield R, Billaut F, Cannon J, Portus MR, Marino FE. Cold-water immersion decreases cerebral oxygenation but improves recovery after intermittent-sprint exercise in the heat. Scand J Med Sci Sports
13. Laptook AR, Shalak L, Corbett RJ. Differences in brain temperature and cerebral blood flow during selective head versus whole-body cooling. Pediatrics
14. Girisgin AS, Kalkan E, Ergin M, et al. An experimental study: does the neuroprotective effect increase when hypothermia deepens after traumatic brain injury? Iran Red Crescent Med J
15. Metz C, Holzschuh M, Bein T, et al. Moderate hypothermia in patients with severe head injury: cerebral and extracerebral effects. J Neurosurg
16. Erecinska M, Thoresen M, Silver IA. Effects of hypothermia on energy metabolism in mammalian central nervous system. J Cereb Blood Flow Metab
17. CryoHelmet. For head injuries. Ice Your Head
. 2014; [cited 2019 Feb 20]. Available from: http://iceyourhead.com/research/for-head-injuries/
18. Walter A, Finelli K, Bai X, et al. Neurobiological effect of selective brain cooling after concussive injury [Internet]. Brain Imaging Behav
. 2017 [cited 2017 Aug 13]. Available from: http://link.springer.com/10.1007/s11682-017-9755-2
19. Plichta MM, Herrmann MJ, Baehne CG, et al. Event-related functional near-infrared spectroscopy (fNIRS): are the measurements reliable? Neuroimage
20. Bhambhani Y, Maikala R, Farag M, Rowland G. Reliability of near-infrared spectroscopy measures of cerebral oxygenation and blood volume during handgrip exercise in nondisabled and traumatic brain–injured subjects. J Rehabil Res Dev
21. Alvarez JA, Emory E. Executive function and the frontal lobes: a meta-analytic review. Neuropsychol Rev
22. Gualtieri C, Johnson L. Reliability and validity of a computerized neurocognitive test battery, CNS vital signs. Arch Clin Neuropsychol
23. Piland SG, Motl RW, Guskiewicz KM, Mccrea M, Ferrara MS. Structural validity of a self-report concussion-related symptom scale. Med Sci Sports Exerc
24. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand
25. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol
26. Bijur PE, Silver W, Gallagher EJ. Reliability of the visual analog scale for measurement of acute pain. Acad Emerg Med
27. Ferrari M, Quaresima V. A brief review on the history of human functional near-infrared spectroscopy (fNIRS) development and fields of application. Neuroimage
28. Matcher SJ, Cope M, Delpy DT. In vivo measurements of the wavelength dependence of tissue-scattering coefficients between 760 and 900 nm measured with time-resolved spectroscopy. Appl Optics
29. Duncan A, Meek JH, Clemence M, et al. Optical pathlength measurements on adult head, calf and forearm and the head of the newborn infant using phase resolved optical spectroscopy. Phys Med Biol
30. NeuroImaging Tools & Resources Collaboratory. NITRC: Homer2: Tool/Resource info. Homer2
. [date unknown]; [cited 2019 Feb 20]. Available from: https://www.nitrc.org/projects/homer2/
31. Mehagnoul-Schipper DJ, van der Kallen BF, Colier WN, et al. Simultaneous measurements of cerebral oxygenation changes during brain activation by near-infrared spectroscopy and functional magnetic resonance imaging in healthy young and elderly subjects. Hum Brain Mapp
32. Jackson K, Rubin R, Van Hoeck N, Hauert T, Lana V, Wang H. The effect of selective head-neck cooling on physiological and cognitive functions in healthy volunteers [Internet]. Transl Neurosci
. 2015;6(1). [cited 2017 Feb 8]. Available from: http://www.degruyter.com/view/j/tnsci.2015.6.issue-1/tnsci-2015-0012/tnsci-2015-0012.xml
33. Plenger P, Krishnan K, Cloud M, Bosworth C, Qualls D, Marquez de la Plata C. fNIRS-based investigation of the Stroop task after TBI. Brain Imaging Behav
34. Zhang ET, Inman CB, Weller RO. Interrelationships of the pia mater and the perivascular (Virchow–Robin) spaces in the human cerebrum. J Anat
35. Bruner E, Mantini S, Musso F, De La Cuétara JM, Ripani M, Sherkat S. The evolution of the meningeal vascular system in the human genus: from brain shape to thermoregulation. Am J Hum Biol
36. Sindelar B, Bailes J, Sherman S, et al. Effect of internal jugular vein compression on intracranial hemorrhage in a porcine controlled cortical impact model. J Neurotrauma
37. Myer GD, Yuan W, Barber Foss KD, et al. Analysis of head impact exposure and brain microstructure response in a season-long application of a jugular vein compression collar: a prospective, neuroimaging investigation in American football. Br J Sports Med
38. Smith DW, Bailes JE, Fisher JA, Robles J, Turner RC, Mills JD. Internal jugular vein compression mitigates traumatic axonal injury in a rat model by reducing the intracranial slosh effect. Neurosurgery
39. Turner RC, Naser ZJ, Bailes JE, Smith DW, Fisher JA, Rosen CL. Effect of slosh mitigation on histologic markers of traumatic brain injury: laboratory investigation. J Neurosurg