Neurologic Assessment of the Neurocritical Care Patient: The

11 Sedation is a ubiquitous practice in ICUs and NCCUs. It has the benefit of reducing 12 cerebral energy demands, but also precludes an accurate neurologic assessment. Because of this, 13 sedation is intermittently stopped for the purposes of a neurologic assessment, which is termed a 14 neurologic wake-up test (NWT). NWTs are considered to be the gold-standard in continued 15 assessment of brain-injured patients under sedation. NWTs also produce an acute stress response 16 that is accompanied by elevations in blood pressure, respiratory rate, heart rate, and ICP. 17 Utilization of cerebral microdialysis and brain tissue oxygen monitoring indicates that this is not 18 mirrored by alterations in overall cerebral metabolism, and seldom affects oxygenation. The hard 19 contraindications for the NWT are preexisting intracranial hypertension, barbiturate treatment, 20 status epilepticus, and hyperthermia. However, hemodynamic instability, sedative use for 21 primary ICP control, and sedative use for severe agitation or respiratory distress are considered 22 significant safety concerns. Despite ubiquitous recommendation, it is not clear if additional 23 clinically relevant information is gleaned through its use, especially with the contemporaneous 24 utilization of multimodality monitoring. Various monitoring modalities provide unique and 25 pertinent information about neurologic function, however, their role in improving patient 26 outcomes and guiding treatment plans has not been fully elucidated. There is a paucity of 27 information pertaining to the optimal frequency of NWTs, and if it differs based on type of 28 injury. Only one concrete recommendation was found in the literature, exemplifying the 29 uncertainty surrounding its utility. The most common sedative used and recommended is 30 propofol because of its rapid onset, short duration, and reduction of cerebral energy 31 requirements. Dexmedetomidine may be employed to facilitate serial NWTs, and should always 32 be used in the non-intubated patient or if propofol infusion syndrome (PRIS) develops. 33 Midazolam is not recommended due to tissue accumulation and residual sedation confounding a 34 reliable NWT. Thus, NWTs are well tolerated in most patients and remain recommended as the 35 gold-standard for continued neuromonitoring. Predicated upon one expert panel, they should be 36 performed at least one time per day. Propofol or dexmedetomidine are the main sedative choices, 37 both enabling a rapid awakening and consistent NWT. 38 Introduction 39 Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 7 December 2020 doi:10.20944/preprints202012.0149.v1 © 2020 by the author(s). Distributed under a Creative Commons CC BY license. The Neurologic Wake-Up Test 2 There is widespread use of sedation for patients in the intensive care unit (ICU) and 40 neurocritical care unit (NCCU). This is a necessary practice to facilitate endotracheal intubation 41 and mechanical ventilation, however, it is also Janus-faced. There is good clinical utility, such as 42 controlling patient distress, attenuating anxiety, and abating pain recognition (1), with neuro43 specific benefits including reduced metabolic demands to decrease energy consumption (2), 44 decreased stress-related injury, as well as seizure, temperature, and intracranial pressure (ICP) 45 control (3, 4). However, over sedation harbors complications, including increased morbidity (5, 46 6), prolonged ventilation with associated pneumonia (3), greater muscular atrophy, venous stasis, 47 thrombosis, and a protracted ICU length of stay. Further, it may increase hospital costs 48 secondary to ordering unnecessary neuroimaging (1). Too little sedation can magnify agitation 49 and autonomic instability, leading to elevated ICP, hypertension, tachycardia, and cerebral 50 oxygen consumption (1). Thus, risks and benefits must be carefully weighed when it comes to 51 achieving optimal sedation. 52 Sedation also precludes an accurate neurologic examination, and continued sedation may 53 mask significant changes in the patient’s neurologic condition (7). This is concerning, as 54 upwards of 40% of patients with a traumatic brain injury (TBI) demonstrate a significant 55 deterioration of neurologic function during the first 48 hours (3, 8). There is a known secondary 56 deterioration that evolves during the early period of brain injury that is heterogenous and hard to 57 predict (9). This is due to secondary injury cascades that activate inflammatory, excitotoxic, 58 metabolic, and vascular phenomena. This augments oxidative stress, elevates ICP and metabolic 59 demands, causes cerebral edema, activates coagulation cascades, and impairs regional blood flow 60 (7). Continued sedation can also prevent the acquisition of an accurate Glasgow coma scale 61 (GCS) score. This is imperative, as GCS scores are a robust prognostic marker and indicator of 62 potential surgical intervention (10, 11), and are highly predictive of 6-month outcomes in TBI 63 patients (12). This underlies the necessity for a brief cessation of sedation for an accurate 64 neurologic assessment, termed the neurological wake-up test (NWT). The NWT is considered to 65 be the gold-standard for neuro-monitoring (1, 3), and is the basis for neuroanatomical 66 localization of pathology, identifying undiagnosed neurologic ailments, detecting early 67 neurologic signs of insult, determining prognosis, and guiding appropriate therapy (3, 7, 13). 68 Serial NWTs are an integral part of continued ICU and NCCU assessment of neurologic 69 functioning but are concerning because they require a temporary cessation of sedation. This 70 results in a significant sympathetic nervous system (SNS) discharge, potentially resulting in 71 neurologic injury via elevating ICP, increasing cerebral oxygen demand, and reducing cerebral 72 perfusion. However, this must be weighed against the additional clinical information acquired. 73 Moreover, with the increasing utilization of multimodality monitoring, there may be enough 74 information gathered to make the NWT both harmful and redundant. Therefore, this paper will 75 aim to elucidate the utility of the NWT, if it still has a role with multimodality monitoring, if 76 there is an optimal frequency that imparts the most favorable risk to benefit profile, and if the 77 choice of sedative has an influence on these quandaries. For this review, PubMed was searched 78 for all existing literature using the terms brain injury, head injury, or TBI, with the terms 79 sedation cessation, daily interruption of sedation, wake-up test, stopping sedation, spontaneous 80 awakening trial, neurologic examination, multimodality monitoring, frequency of awakening, 81 and/or frequency of neurologic exam. 82 Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 7 December 2020 doi:10.20944/preprints202012.0149.v1


83
The brain is known to be the most metabolically active organ in the body by weight, with 84 a cerebral metabolic demand for O2 (CMRO2) between 3-3.5 ml O2/100g/min, which is roughly 85 15-20% of total cardiac output. To maintain this high demand for cerebral blood flow (CBF), 86 adequate cerebral perfusion pressure (CPP) is required, defined as the difference between mean 87 arterial pressure (MAP) and ICP. In a healthy brain, there is robust autoregulatory mechanisms to 88 maintain constant CBF across a range of MAPs from 65-150 mmHg. However, in the presence 89 of neurologic insult, there is often either regional or global impairment of autoregulation (1). 90 Thus, continued monitoring of ICP and CPP are often necessary. Utilizing transcranial doppler 91 (TCD) may also assist in assessing the degree of autoregulation failure (14). Other monitoring 92 modalities include brain tissue oxygen tension monitoring (PbtiO2), jugular venous oxygen 93 saturation (SjvO2), and brain neurochemistry by intracerebral microdialysis (MD). Newer, less 94 invasive monitoring technologies including optic ultrasound, and the automated pupillometer to 95 directly augment the NWT are also seeing increased use.

97
Normal ICP values are between 5-15 mmHg (15), with a generally accepted normal value 98 of <20 mmHg. Increased ICP is defined as pressures greater than 20 mmHg for more than 5 99 minutes. This monitoring is typically performed via insertion of a catheter into the lateral 100 ventricle, acting as an external ventricular drain (EVD), which concomitantly allows for 101 cerebrospinal fluid (CSF) drainage for ICP control if required (7). ICP monitoring is 102 recommended as part of the official TBI guidelines (16); consequently, there is widespread use 103 of ICP monitoring of brain injured patients across NCCUs (7,9,17). The Brain Trauma 104 Foundation recommends monitoring for comatose patients (GCS of 3-8) that have an abnormal 105 CT scan (18). Those with the highest risk of developing intracranial hypertension following head 106 injury are severely head injured comatose patients (GCS <8), with very few patients with mild to 107 moderate head injury showing secondary deterioration (18). 108 Increased ICP is a recognized cause of morbidity and especially mortality after TBI (7, 109 12, 18-20). Mechanistically, elevated ICP can acutely cause ischemia by directly reducing CBF. 110 ICP is frequently elevated following neurologic insult, compounded by a failure of CBF 111 autoregulation, which exacerbates the problem (9). There is evidence to suggest that aggressive 112 management of elevated ICP can improve outcomes in TBI patients (9,18). Conversely, 113 concerns have been raised regarding no improved clinical outcomes with ICP monitoring, and 114 possibly increased mortality rates from its use, at least in TBI patients (7,21). A recent Cochrane 115 review corroborated this, finding insufficient evidence for routine ICP monitoring in TBI 116 patients (22). The major study considered in this review was the BEST-TRIP trial (23), which 117 demonstrated that treatment guided by ICP-monitoring was not superior to treatment guided by The purpose of SjvO2 monitoring is to acquire information pertaining to cerebral oxygen 147 supply, perfusion, and consumption. It can be monitored either via fiberoptic catheters placed 148 into the internal jugular bulb immediately distal to the jugular foramen, or intermittently 149 checking jugular venous blood samples (7,9). A noteworthy technical hurdle also exists, as 150 falsely elevated SjvO2 can be measured due to the presence of only extracerebral blood in the 151 jugular bulb (29). This depends on contamination from extracerebral sources via aspirating too 152 quickly, or misplacing the catheter by only a couple of centimeters (30), which may lead to 153 erroneous results and mismanagement (9). Moreover, this modality is of limited utility in severe 154 global ischemia or very large infarcts, as SjvO2 can rebound upward due incomplete oxygen 155 extraction by the ischemic tissue (29, 31).

156
Normal values range from 55-75% oxygen saturation (O2 sat). At baseline, shifts in detected (9, 35), prompting unnecessary workup. Its role in monitoring patients in the NCCU has 168 not yet been determined (7) Direct monitoring of brain tissue oxygenation provides a continuous stream of 172 information pertaining to focal oxygenation, with typical values ranging from 15-30 mmHg (2-4 173 kPa), and a critical ischemic threshold commonly established at 10 mmHg (<1.33 kPa), which is 174 indicative of brain hypoxia (7,9). This measurement is carried out through a thin electrode 175 placed into white matter, either in peri-ischemic at-risk tissue (the penumbra) for focal 176 measurements, or in frontal white matter to estimate global cerebral oxygenation in diffuse brain 177 injury (7,9). Using this measure, ischemic changes with regional differences have been detected 178 following TBI (36), and these transient periods of ischemia are correlated with worsened patient 179 outcomes (37). In general, brain injured patients with brain hypoxia (PbtiO2 <10 mmHg) have 180 significantly poorer outcomes and increased mortality (38) combinations of multimodality monitoring are more effective and provide additional insight not 190 garnered by either individually (7). Despite strong correlations between brain hypoxia and poor 191 outcomes, and retrospective reviews suggesting some benefit to treatments guided by PbtiO2 192 measurements, high quality randomized trials are required to determine if guiding therapeutic 193 interventions based upon this metric is beneficial (12,25,42). Consistently, PbtiO2-guided 194 therapy and clinical outcomes remain subject to debate (42). 196 Intracerebral MD can be utilized to measure brain neurochemistry through the insertion 197 of a microdialysis catheter that contains a semipermeable hollow fiber membrane at its tip. This 198 is attached to a perfusion pump that is subsequently perfused with a fluid that resembles the 199 interstitial fluid in question (artificial CSF). This allows transport of molecules across a 200 concentration gradient through the membrane via passive diffusion, allowing measurement of 201 various chemicals, such as neurotransmitters, and metabolic substrates and products, like 202 glucose, lactate, pyruvate, glycerol, glutamate, et cetera (7). The MD catheter is typically placed 203 adjacent to a focal lesion to detect early metabolic alterations, such as near a vascular territory 204 susceptible to vasospasm following subarachnoid hemorrhage (SAH), or the penumbra region of 205 an infarct or mass lesion. It may also be placed in the non-dominant frontal region in the case of 206 diffuse injury (9, 43).

207
MD measures the cellular bioenergetic and redox state of the monitored brain region, 208 which is typically perturbed in brain injury. Ischemia and resultant lack of blood flow and 209 oxygenation shifts metabolism from the brain's obligatory aerobic condition to an anaerobic one, 210 resulting in a lack of ATP production. (44). This is compounded by severe mitochondrial 211 dysfunction and inhibition of pyruvate dehydrogenase (PDH) in TBI patients (45,46), causing 212 inhibition of oxidative phosphorylation. This results in a surge in lactate levels as a means to 213 restore NAD+ levels, reflected by a high lactate to pyruvate ratio (LPR) (44,46). Glutamate is 214 elevated post-brain injury for two reasons: ATP is necessary to maintain baseline ionic 215 equilibrium across various channels; thus, its absence causes massive glutamate release due to 216 postsynaptic calcium entry, and glutamate reuptake and metabolism are ATP-dependent 217 processes (44, 46). Glucose levels plummet due to either lack of blood flow, or glucose 218 hyperutilization following TBI, both of which can reduce glucose concentration below critical 219 thresholds (25). Glycerol elevation indicates decomposition of cell membranes associated with 220 severe cellular damage (43). Correlating the microscopic anomalies with the macroscopic 221 picture, these processes cause neuronal death, cellular edema, and elevated ICP (47).

222
As MD detects metabolic alterations, it can provide valuable clinical information about 223 brain function following injury and impending metabolic crises. Its most promising application is 224 detecting ischemia and neuronal damage at pre-clinical stages, allowing for early intervention to 225 salvage brain tissue (25,47). LPR is a sensitive marker of brain ischemia and redox state (9, 43).

226
Elevated LPR measurements correlate with symptom severity and fatal outcomes after brain 227 injury (43, 47). Elevated LPR >25 is associated with poor outcomes in TBI (9), and elevated 228 LRP coupled with low glucose correlates with worsened outcomes in TBI and SAH patients (25, 229 48). Furthermore, one study showed that length of time spent with elevated LPR >40 correlated 230 with frontal lobe atrophy at 6 months (49).

231
Additionally, MD is purported to aid in guiding optimal CPP targets (25, 47), although 232 this has been inconsistent (50). Elevated levels of glutamate show similar predictive power as a 233 proxy for ischemia (9, 43), and the degree of elevated glutamate correlates with poor patient 234 outcomes (47, 51). However, this has been challenged by some (47, 52). Elevating glycerol 235 levels ≥ 3 days post TBI can be associated with ongoing neurologic deterioration or seizure 236 activity (9, 47). Thus, MD offers unique insight into cellular bioenergetics and their perturbations 237 following brain injury, possibly before clinical signs could be elicited on examination. There is 238 increasing use of MD, and certain protocols have been established, with alarm levels of LPR set 239 at >30, and/or glucose levels < 0.8 mmol/l (53). Despite the promising utility, its overall value as 240 a tool for guiding clinical decision making has yet to be fully elucidated (7,25,53 Medicine held a panel to evaluate and discuss the evidence of multimodality monitoring (12).

257
They conclude that a single monitoring modality is demonstrably insufficient. Despite this, they 258 state that there is no consensus on their use, and more studies are required to ascertain if it 259 translates to improved patient outcomes. However, they underscore the importance of the 260 neurologic exam and the NWT, stating that it remains a cornerstone in the accurate assessment of 261 patients. The vast amounts of information gathered via these monitoring modalities, how to 262 evaluate and integrate them, and their ability to guide optimal therapeutic plans is still being 263 elucidated (10).  Evolving pathology can be picked up earlier, and some deterioration may only be detected by 301 physical examination (7). This helps in patient assessment, and in monitoring treatment perturbations are well tolerated, and are unlikely to cause secondary injury to the brain (7, 59).

360
In one prospective study of 20 patients (54 total NWTs), (78), upwards of 34% of NWTs 361 were not performed, as the patients were not seen as stable enough due to elevated ICP and need 362 for continuous sedation. In those that received NWTs, upwards of 33% were stopped due 363 primarily to ICP-crisis (>20 mmHg), or systemic desaturation. Moreover, in those patients with 364 NWT cessation, they noted a statistically significant reduction in PbtiO2 measures, decreasing 365 from an average of 28 mmHg to 19 mmHg, which is still well above hypoxic thresholds (9, 79).

366
While not rising to the level of statistical significance, a trend was noted between lower brain 367 glucose, higher total LPR, and neurologic deterioration from the NWT, which seems to 368 corroborate the benefit of multiple simultaneous modalities of monitoring. Overall, however, 369 there were no significant alterations in lactate, pyruvate, LPR, or glucose in patients undergoing 370 NWTs. Moreover, as shown by Skoglund et al (63,77), transiently increased ICP to >20 mmHg 371 may not be as deleterious as presumed. If they are not coupled with concomitant metabolic 372 indicators of ischemia or inadequate perfusion, it is unlikely that neurologic deterioration is 373 occurring; in no studies heretofore has the NWT been shown to cause neurologic injury.

374
Therefore, though patients that developed intracranial hypertension in response to the NWT had 375 stoppage of the examination, the preponderance of data indicate most of these patients were 376 likely to tolerate the exam without promoting neurologic injury. Nevertheless, differential 377 findings in PbtiO2 measures between this report by Helbok et al. (78) and the aforementioned 378 one by Skoglund et al. (77), further demonstrate the marked heterogeneity of brain injured

416
The concept of "ICP dose" is becoming increasingly investigated. It recognizes that ICP 417 thresholds are arbitrary and meaningless without accounting for the time spent at "deleterious"  The NWT remains contemporaneously regarded as the sine qua non for optimal 451 assessment of the brain-injured patient (3,7,12,13,70,78 handled in a case-by-case basis with appropriate risk-benefit profiles appraised.

460
The reliability of the neurologic exam, its cost, and its ability to detect subtle deficits not  Overall, multimodality neuromonitoring should be seen as a compliment to the NWT, and vice versa (55). More investigations are required to tease out the absolute clinical utility of the NWT, 469 with specific regard to patient outcomes and management guidance (60).

470
Is there an optimal frequency of performing NWTs? 471 Another central question pertaining to NWTs is their optimal frequency of utilization. with ongoing ICP monitoring, but do not mention frequency (66). One report suggests that 478 upwards of 50% of NCCU centers in Scandinavian countries do not utilize NWTs (92), which 479 may partly be explained by differences in sedative use. In those centers utilizing NWTs, the 480 majority use daily, with sometimes twice daily checks (7, 59), and one center utilizes NWTs 481 between 4-6 times per day (7).

482
This demonstrates the enormous variability in both use and frequency of NWTs, 483 necessitating the need for more data and guidelines to be established to guide clinical practice.

484
Indeed, only one official recommendation has been offered, from the European Society for 485 Intensive Care Medicine (ESICM) NeuroIntensive Care Section (NIC) (13). They convened an 486 expert panel and state that a daily, brief interruption of sedation is recommended to facilitate an 487 accurate neurologic examination (an NWT) and improve outcomes, giving it moderate evidence 488 and a strong recommendation. Additionally, they state that brain injured patients, ICU patients, 489 and in general all critically ill patients, shoulder undergo a neurological examination on initial 490 ICU admission, and at least once daily, giving it moderative evidence, and a best practice 491 recommendation. They also note the clear contraindication, as previously stated, with DIS and

492
NWTs not recommended in patients with preexisting intracranial hypertension, assigning it 493 moderate evidence and a strong recommendation (13). Lastly, in their concluding remarks, they 494 note that despite technological advances, the neurologic examination remains a foundation of 495 accurate evaluation and prognostic assessment of neurologic function, pointing out the robust 496 prognostic value of GCS scores and pupillary light responses. Therefore, the only concrete 497 recommendation that has been put forth pertaining to the optimal frequency of NWTs is at 498 minimum once daily. Randomized controlled trials will need to be undertaken utilizing 499 differential NWT frequencies to gain insight into what the optimal frequency is, if different 500 patient populations have an impact on this, and what influence injury type has. Thus, much more 501 research will need to be carried out before concrete recommendations about NWT frequency can 502 be given and guidelines administered.

504
Another consideration is sedative selection. There are myriad sedating agents utilized in 505 the ICU and NCCU, namely propofol, benzodiazepines, dexmedetomidine, opioids, and 506 barbiturates. The most common agents utilized in ICUs and NCCUs are propofol and midazolam 507 (7), and these may be used in conjunction with opioids for additional analgesia. Recently, 508 dexmedetomidine has begun to see increasing use, and it represents another attractive option for 509 sedation. They each possess their own risk vs benefit profiles.

510
Propofol enjoys ubiquitous ICU and NCCU use owing to several factors, including its 511 neuroprotective effects. It is recommended to use for ICP control in current TBI-care guidelines 512 (16), and it dampens cerebral metabolic oxygen demand (89). It possesses both rapid onset of 513 action, and rapid plasma clearance, which facilitates reliable recovery of consciousness even 514 after prolonged administration and thereby a consistent NWT. In higher doses it can induce burst 515 suppression, which can effectively treat status epilepticus (89). It can abate oxidative stress, 516 making it especially useful to combat the free radical generation in head injury (93). A 517 microdialysis study in TBI patients comparing propofol to midazolam found no significant 518 differences in measures of LPR, glutamate, glycerol, or glucose between the two agents over a 519 72-hour period (94). The doses utilized in that study may have been inadequate, and it was a 520 short-term, small study, therefore more investigations are required to determine if propofol can  (93). Propofol is an ideal agent for the NWT, and is widely 537 recommended and utilized. Although rare, PRIS may limit its utility for long-term use in brain 538 injured patients.

539
Midazolam has rapid onset with a rapid recovery. It decreases cerebral oxygen demand, 540 although to a lesser extent than propofol, and has only slight influences on ICP levels (89).

541
Despite possessing a short half-life, when infused continuously, its half-life increases due to both 542 high lipid-solubility with associated tissue accumulation, and the presence of active metabolites 543 that may be deleterious (7, 89). Consistently, midazolam use is associated with higher mortality 544 rates in ICUs (98). Its persistent use leads to protracted sedation and prolonged time to 545 awakening, which will confound a consistent NWT (89,99). It also causes significant respiratory 546 depression and inhibition of the cough reflex, along with issues of tolerance development and 547 significant withdrawal symptoms upon cessation. Its use is also a risk factor for ICU delirium, 548 which is itself associated with worse outcomes (93). Accordingly, its use increases ICU length of 549 stay, and either propofol or dexmedetomidine sedation is preferred to improve clinical outcomes 550 in intubated ICU patients (100). Midazolam is not recommended for use when serial NWTs are 551 desired.

552
Dexmedetomidine is a sympatholytic agent that can achieve excellent sedation without 553 respiratory depression, and possesses anxiolytic and analgesic properties. It has rapid onset of 554 action and elimination, does not accumulate in tissues, has a half-life of 6 minutes, and a 555 terminal elimination half-life of 2-2.5 hours, making it ideal for a reliable NWT (89,93,101). It 556 also decreases ICP through reducing CBF, increases CPP, and reduces incidence of delirium 557 (93). In one report, dexmedetomidine alone versus propofol alone showed no significant 558 differences in measures of ICP, CPP, tachycardia, bradycardia, hypertension, or length of stay 559 (102). Another report demonstrated that both midazolam and propofol increased the number of 560 ventilator-associated events, whereas dexmedetomidine did not (103). Dexmedetomidine was 561 associated with increased chance of extubation in that study. In a TBI murine model, 562 dexmedetomidine showed significant neuroprotective effects (104), although human studies are 563 required to corroborate these findings. Given its known role of impeding SNS discharge (105), it 564 may decrease the injury-promoting catecholaminergic influence in TBI (106), and has been 565 shown to decrease plasma cortisol after administration (107). This could make this choice of 566 sedative especially useful in the NWT for lessening the NE and E excursions. Moreover, it has 567 been shown that in comparison to propofol or midazolam, dexmedetomidine patients were better 568 able to be aroused and cooperate, suggesting it may facilitate a more appropriate level of 569 sedation, better enabling serial NWTs (108). Overall, this agent represents an extremely 570 attractive option to utilize for long-term sedation and to facilitate an NWT. However, more 571 studies are required, and given its limited clinical data, some authors do not currently 572 recommend its use for sedation in brain-injured patients (89).

573
Most recommendations call for continuous propofol sedation, allowing for more frequent 574 cessation of sedation for NWTs (3, 7, 60, 93). There has been some consideration of maintaining 575 a low-dose of analgesics during NWTs (60). Thus, the few recommendations on the subject 576 suggest using propofol to facilitate a smoother transition toward the NWT, with careful attention 577 paid to the development of PRIS, and ensuring infusion rates remain less than 4 mg/kg/hour 578 unless for bolus ICP control. Dexmedetomidine has not yet received strong recommendations, 579 especially in the NCCU, due to an absence of large-scale data. However, the existing literature 580 indicates that dexmedetomidine can be safely and efficaciously used for brain-injured patients, 581 and its rapid onset and short half-life devoid of residual tissue accumulation make it a very 582 attractive choice to facilitate serial NWTs. It is already utilized in NCCUs that employ NWTs 583 and has established feasibility and tolerability for the process (78). It is strongly recommended as 584 the sedative agent of choice in the non-intubated patient (93).

585
Propofol is most strongly recommended due to widespread utilization, existing clinical 586 research, and the preponderance of recommendations espousing its use. Despite this, 587 dexmedetomidine can likely be used in its place. The primary choice of sedative should be 588 individualized, and further based on comfort levels and experience. In non-intubated brain 589 injured patients, or with the development of PRIS, dexmedetomidine should be the primary 590 sedative. Midazolam use is not recommended for sedation because it precludes the ability to 591 perform reliable, serial NWTs (93); it is associated with greater ICU length of stay, duration of 592 mechanical ventilation, and delirium compared to propofol or dexmedetomidine (109); and is 593 associated with dramatically prolonged time to awakening compared to propofol (110-113) and 594 dexmedetomidine (114,115). Despite the numerous pitfalls of midazolam use, it remains one of 595 the most common sedatives utilized in ICUs (3). Therefore, in centers that primarily use 596 midazolam, it must be recognized that NWTs may lack consistency due to bioaccumulation and 597 residual sedation, with much greater time to awakening. Notwithstanding, when multimodality 598 monitoring is not utilized the NWT becomes the only source of information about neurologic 599 function. This is already a reality in many lower-income countries and areas, with assessment 600 coming from neurologic exams and serial CT imaging (116). In those cases, the NWT should 601 still be done despite midazolam use, with acknowledgement that time to awakening is increased 602 and day to day reliability of the exam diminished.

604
There are myriad benefits of sedation in the ICU and NCCU, but the most prominent in 605 neuro-critical care is the reduction in cerebral metabolic demand. It is also widely recognized 606 that the NWT is considered the gold-standard for evaluation of patients. Even with ubiquitous 607 endorsement, its use cannot be justified if it does not provide additional clinically relevant 608 information and guide decision making. Therefore, more studies are necessary to totally clarify 609 its utility in clinical decision making. term sedation in general, nor as a good agent to facilitate consistent NWTs. In centers that 632 primarily employ midazolam, there must be an appreciation for an inability to perform 633 consistent, serial NWTs in a timely manner.

634
In patients without multimodal monitoring, the NWT should be utilized provided no hard 635 contraindications exist. Additionally, vital signs must be stable before and throughout the 636 examination (especially O2 saturation). The NWT should not be used if sedation is being used as 637 a primary treatment for elevated ICP, severe agitation, or respiratory distress. Clinical suspicion 638 should be used to identify those at the greatest risk of developing intracranial hypertension, and 639 risk factors include an abnormal CT scan on admission, hypotension on admission, GCS-M < 3, pupillary anomalies, or a lengthy inability to follow the exam (10). Multimodal monitoring can 641 compensate for the inability to perform an NWT when contraindicated. Currently, based upon 642 one expert panel, NWTs can be recommended at a minimum of once per day. Either propofol or 643 dexmedetomidine represent good options for sedation. More research is required to elucidate the 644 clinical utility of the NWT, and establish guidelines about the optimal frequency of its utilization 645 with stratification based on injury type and patient population.