The cerebrovascular response to norepinephrine: A scoping systematic review of the animal and human literature

Abstract Intravenous norepinephrine (NE) is utilized commonly in critical care for cardiovascular support. NE’s impact on cerebrovasculature is unclear and may carry important implications during states of critical neurological illness. The aim of the study was to perform a scoping review of the literature on the cerebrovascular/cerebral blood flow (CBF) effects of NE. A search of MEDLINE, BIOSIS, EMBASE, Global Health, SCOPUS, and Cochrane Library from inception to December 2019 was performed. All manuscripts pertaining to the administration of NE, in which the impact on CBF/cerebral vasculature was recorded, were included. We identified 62 animal studies and 26 human studies. Overall, there was a trend to a direct vasoconstriction effect of NE on the cerebral vasculature, with conflicting studies having demonstrated both increases and decreases in regional CBF (rCBF) or global CBF. Healthy animals and those undergoing cardiopulmonary resuscitation demonstrated a dose‐dependent increase in CBF with NE administration. However, animal models and human patients with acquired brain injury had varied responses in CBF to NE administration. The animal models indicate an increase in cerebral vasoconstriction with NE administration through the alpha receptors in vessels. Global and rCBF during the injection of NE displays a wide variation depending on treatment and model/patient.


| Search results and study characteristics
The results of the search strategy across all databases and reference sections of articles are summarized in Figure 1. Overall, a total of 2463 articles were identified, all from the databases searched.
A total of 635 were removed because of duplication of references, leaving 1825 to review. By applying the inclusion/exclusion criteria to the title and abstract of these articles, we identified 288 articles that fit these criteria. Six articles were added from reference sections of pertinent review articles, leaving a total of 294 papers to review. The portable document formats (PDFs) of these 294 were then gathered. Applying the inclusion/exclusion criteria to these PDFs, only 88 articles were found eligible for inclusion in the systematic review. Articles were excluded because they either: did not report F I G U R E 1 PRISMA flow diagram details around the CBF/cerebrovascular response to NE administration or were nonrelevant. One article was a retrospective study focused on CBF and brain function during hypotension and shock. 16

| NE impact on objectively measured CBF
The following subsections provide a narrative summary of the impact of NE administration on objectively measured cerebrovascular response/CBF, looking first at overall increase/decrease in CBF, followed by measured models pathology-specific responses to NE.

| Increase in CBF
Twenty-six studies demonstrated an increase in global or rCBF with the administration of NE. 17 76 The CBF increase ranged from not significant, to changes on the order of 500% of the initial CBF value. 45 In studies which measured rCBF, all areas increased in blood flow except the auditory cortex 65 and mesencephalon. 62 Six studies had a dose-dependent increase in CBF, 40,45,46,[70][71][72] with one study showing a peak CBF at a NE dose of 0.16 mg/kg. 45 The variation within the data between the individual animal models and pathologies limits the ability to draw any clear conclusions within species or technique.

| Direct vascular response
Of the seven studies that evaluated direct cerebrovascular response to NE, 20,26,27,52,53,59,77 all had some form of constriction to the cerebral vessels. This cerebral vessel change ranged from nonsignificant up to 20% constriction as compared to baseline values. 52,53,59 However, models that had a significant constrictive response to NE were either injected with another solution (a hypertonic saline solution or Wahl solution 20 ) or had NE locally applied to cerebral vessels. 52,53,59
First, the influence of coadministered substances on the effects of NE demonstrated some findings of interest. Such substances include: hypertonic saline, 18,62 phentolamine, 24,33,59,61 phenoxybenzamine, 32,64 and propranpol. 31,32,49,62 Hypertonic saline injected with NE significantly increased CBF and cerebral metabolic rate of oxygen consumption (CMRO 2 ) as compared to NE alone. 18,62 Similarly, when NE was allowed to pass the blood-brain barrier (BBB) after osmotic opening with urea, an increased regional flow was obtained. 62 Phentolamine inhibited or completely mitigated the CBF TA B L E 1 Included studies-general characteristics and study goals Finally, in one study, the constriction of large arterioles was induced through NE, with pial vessels remaining unchanged. 20 While in rats, the carotid blood flow was decreased by 0.5 mL/min in all ages of animals with the injection of NE. This study also found the Propranolol + isoprenaline: Clear-cut increases in regional blood flow were found in pons, mesencephalon, thalamus, and caudate nucleus.
The cortical regions and cerebellum only showed a tendency to flow increase, which was not statistically significant

None mentioned
The presence and heterogenous distribution in the cerebrovascular bed of alpha-and betaadrenoceptors that can be activated by sympathomimetics given systemically. If NE was allowed to pass the blood-brain barrier after osmotic opening with urea, an increased regional flow was obtained, probably due to a mechanism where the vasodilator effect secondary to activation of cerebral metabolism predominated

TA B L E 2 (Continued)
carotid vascular conductance was different with 0.005 mL/min in juveniles, and 0.08 mL/min in mature and middle-aged rats, suggesting an age-related disparity in CBF modulation. 73

| Models given cardiopulmonary resuscitation
There were four studies in pigs that evaluated CBF while CPR was administered. [44][45][46][47] During CPR, NE was given in two studies at varying doses, resulting in a dose-dependent increase to CBF. 45,46 Furthermore, increases in CMRO 2 and CPP were also shown with the injection of NE. 47 One of these studies had NE co-injected with epinephrine and vasopressin, resulting in a more apparent increase in CBF, than compared to epinephrine or the vasopressin alone. 44 In all of these studies, CBF increased with NE in comparison to control animals where NE was not given, with the NE effect on CBF observed to dissipate after 5 minutes. 44,47

| Models with traumatic brain injuries
In the four studies that had head trauma models, three of them used pigs 48,50,51 and one used rats. 74 In general, TBI caused a decrease in CBF, after the injury NE was given which caused an increase in CBF back to near baseline levels. 50,51,74 The partial pressure of oxygen was also increased in the one study that monitored blood gases. 50 One study compared the CBF effects of NE in brain-injured piglets (fluid percussion injury) vs uninjured pigs. This study showed minor changes in CBF by NE in the uninjured pigs, but a significant increase in CBF by NE in the injuried. 51 In the study with rat TBI models, NE administration led to an increase in ICP for 30 minutes, with a gradual decrease in CPP and slight decrease in CBF. 74

| Other studied pathologies
There were some "other" pathologic states studied, including those with sympathectomy, 55 induced intercranial hypertension, 23,25,58 induced hypothermia, 21,35,66 brain tumors, stereotaxic induced lesions, 35 and endotoxic shock. 43 In the studies that had models with removed ganglion 41,56 or sympathetomy, 55 there was a nonsignificant change tin CBF. However in models with ligated bile ducts NE both decreased CBF and increased CVR as compared to NE alone. 19 Whereas, in dogs with a brain tumor (induced by avian sarcoma virus) NE decreased CBF in both hemispheres (one with tumor/one without) and a subsequent decrease in partial pressure of oxygen. 28 In models with stereotaxic lesions (made in the posterior hypothalamus, unilaterally or bilateral) 35 or endotoxishock 43 there was limited change in CBF or practical pressure of oxygen. 35

| Anesthesia in models
In the identified literature there were six studies where the animal model was not fully anesthetized. [38][39][40][41][42]55,60 Within these studies there was a dose-dependent change CBF seen in these models 40,41 and a constrictive force seen by NE injection. 40

| Adverse events
No human studies document the adverse effect to NE but three animal studies included adverse events. 43,64,71 Two studies reported lethal doses of NE administration. 64,71 In one study, the cause of death was determined to be the inhibition of autoregulation by NE. 64 This study also reported that continuous moderate doses of NE for longer than 2 hours prevented autoregulation measured through autoradiography. 64 In TBI models, there appeared to be a trend toward vasoconstriction and varying global and rCBF reductions with NE administration.

| D ISCUSS I ON
NE is commonly used to treat life-threatening low blood pressure situations for its direct vascular effects. 2  entering the extracellular space. Further investigation is required into the regional disparities of CBF secondary to NE in the context of acquired brain injury.
Furthermore, the injection of NE through systemic routes may have effects different than NE directly injected within the brain.
NE injected with hypertonic urea or MgSO 4 solution resulted in an increase in CBF with the same dose of NE. As such it is likely that the BBB mediates the perfusion of NE throughout the brain and its effects on CBF. 18,40,62 This point may also by enforced by the fact that during studies where animals had lesions that opened the BBB, an increase in CBF after NE injection was seen. 40 Also in studies with impaired autoregulation there was a consistent response to NE with an increase in ICP and CBF. 23,24 All these findings support a potential role for the BBB in the regulation on cerebrovascular response to NE. As mentioned above, in line with this, NE given systemically may not enter the brain parenchyma due to the BBB, though it is clear that the BBB limits the permeation of NE it may not prevent all of the NE from entering the BBB. 18,108 This particular area of BBB integrity, its impact on NEbased cerebrovascular/CBF responses in acquired brain injury, is an area requiring much further investigation.
Third, six studies demonstrated that the exogenous administration of NE reaches a maximal effect on cerebrovascular response. 20,29,31,32,45,71 All of these studies compared various doses of NE which resulted in a maximum change in both CBF and CVR of the animal models. Thus, a dose-dependent response to NE occurs, which again carries important implications for continuous cerebrovascular reactivity monitoring and derivation of individualized physiologic target in TBI. However, a universal max dose of NE, in terms of CVR effect, could not be demonstrated due to the heterogeneity within the studies, and is unlikely to exist in vivo in humans.

NE-dosing thresholds and their impact on continuously monitored
cerebrovascular reactivity/CBF in vivo in critically ill neurological patients, such as the TBI population, have not been conducted, and require further investigation.
Finally, unwanted cerebral physiologic side effects of NE administration were seen. Demonstration of NE's impact on ICP was shown by using an extradural balloon to increase ICP. NE had no effect on the overall ICP, unless ICP was at the extreme pressure of over 70 mmHg. 23 Furthermore in studies that measure CBV and CBF, the data demonstrated a positive linear connection between them. 21,25,55,72 This linear connection encourages the idea that potentially the change in CBF has more to do with alterations in CBV than the CVR effect of NE. The inhibition of autoregulatory hemodynamics within the brain by NE injections was also described. 64 Prolonged or long continuous injections have resulted in lethal inhibition to cerebral hemodynamics, as highlighted in two studies. 64,71 As mentioned above, regarding individualized physiologic targets in TBI care, this aspect of prolonged high-dose NE administration needs to be considered and investigated further.

| Limitations
In this review we have been able to systematically, and compre- Furthermore most human studies measured CBF through an assumption of MCAv, this is not a true measure of global CBF or rCBF.
Another limitation is the lack of blood gas control in some of the studies. Cerebrovascular/CBF physiologic response is intimately linked to pCO 2 and pO 2 status, therefore due to the large number of studies that did not fully account for fluctuations in the blood gas level, leaves any conclusion linked with NE deficient. Last, although there are trends in the animal models, there is still a significant limitation to apply them in clinical practices simply based on the limited number of effect human studies.

| Future directions
Further prospective studies on the cerebrovascular/CBF effects of NE in the neurologically ill patient population need to be performed to determine the role of this medication within neuroanesthesia and the neuro-ICU. The potential CBF trends seen with NE are interesting and carry important implications in the treatment of a variety of cerebral pathologies, with TBI mentioned as exemplar given that CBF and cerebral autoregulation are key factors to improve patient outcome. When it comes to TBI, literature in the field of moderate/severe TBI has demonstrated that impaired cerebral autoregulation/cerebrovascular reactivity is directly associated with poor 6-month global outcome. 6,104,106,109,110 This has been validated in prospective multicenter data, 106 and recent retrospective data sets suggest that cerebrovascular reactivity remains unaffected by changes in guideline-based management of TBI over the last 25 years, in concert with relatively stable mortality rates. 105  Cerebrovascular reactivity monitoring is being adopted to direct personalized physiologic targets in TBI care, including optimal CPP targeting, 10,13,112 with the expectation that such personalized approaches based on cerebrovascular monitoring will be extrapolated to other neuropathological states. 7,113-117 Such concepts are currently being explored in phase II clinical trials. 13 Thus, knowledge of the impact of commonly administered vaso- In addition to this evaluation of NE in TBI with advanced multimodal monitoring of cerebrovascular response, further animal models are required. As seen in the described literature body, the presence of TBI or other acquired brain injury may lead to different CBF responses compared to healthy animals/humans. This suggests a potential role for the regional disparities in BBB integrity mitigating the cerebrovascular/CBF response to NE. Future work into NE models with BBB disruption is required to provide insight into the impact of BBB integrity on NE effects. In parallel to this, the control mechanisms involved in cerebral autoregulation are multifaceted, and are likely the impact of individual genetic polymorphisms in humans. 118,119 Future work, both in humans and genetically controlled animal models may also shed insight into variances in cerebrovascular/CBF catecholamine responses. All such work mentioned requires substantial coordination between multiple centers of excellence/expertise, and requires multidisciplinary research teams.
This is the focus of ongoing collaborative work in Europe 120,121 and Canada. 122

| CON CLUS IONS
The animal models indicate an increase in vasoconstriction with NE administration through the alpha receptor in vessels. There appeared to be a dose-dependent increase in CBF with NE administration in healthy and CPR animal models, which was also seen in one human study. However, there was no clear trend to describe the global and rCBF changes seen during the injection of NE in models with TBI, acquired brain injury, or within any other group of human patients.
Further investigation into the impact of NE on cerebrovasculature in large animal models and humans is required.

D I SCLOS U R E
There is no conflict of interest by any of the authors in the work presented.