UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Carbon dioxide mediated vasomotion of extra-cranial cerebral arteries : a role for prostaglandins? Hoiland, Ryan Leo 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_september_Hoiland_Ryan.pdf [ 5.11MB ]
Metadata
JSON: 24-1.0166400.json
JSON-LD: 24-1.0166400-ld.json
RDF/XML (Pretty): 24-1.0166400-rdf.xml
RDF/JSON: 24-1.0166400-rdf.json
Turtle: 24-1.0166400-turtle.txt
N-Triples: 24-1.0166400-rdf-ntriples.txt
Original Record: 24-1.0166400-source.json
Full Text
24-1.0166400-fulltext.txt
Citation
24-1.0166400.ris

Full Text

CARBON DIOXIDE MEDIATED VASOMOTION OF EXTRA-CRANIAL CEREBRAL ARTERIES: A ROLE FOR PROSTAGLANDINS?   by   Ryan Leo Hoiland   B.H.K., The University of British Columbia, 2013   A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF   MASTER OF SCIENCE   in   THE COLLEGE OF GRADUATE STUDIES   (Interdisciplinary Studies)     THE UNIVERSITY OF BRITISH COLUMBIA   (Okanagan)   July 2015     ©Ryan Leo Hoiland, 2015    ! ii!Abstract  Cerebrovascular regulation during perturbations in arterial CO2 is thought to occur solely at the level of the pial vessels. However, recent evidence implicates large extra-cranial cerebral blood vessels in this regulatory process. Although the mechanisms governing CO2 mediated vasomotion remain unclear, animal and human studies support a large role of prostaglandins. Thus, we examined two hypotheses: 1) vasomotion of the internal carotid artery (ICA) would occur in response to both hyper and hypocapnia; and 2) pharmacological inhibition of prostaglandin synthesis with Indomethacin (INDO; a non-selective cyclooxygenase inhibitor) would reduce the vasomotor response of the ICA to changes in end-tidal PCO2 (PETCO2). Using a randomized single-blind placebo controlled study, subjects (n=10) were tested on two occasions. Before and 90-minutes following either oral INDO (1.2mg/kg) or placebo capsule, concurrent measures of beat-by-beat blood flow, velocity and diameter of the ICA were made at rest and during steady state stages (4 min) of iso-oxic hypercapnia (+3, +6, +9mmHg above baseline) and hypocapnia (-3, -6, -9mmHg below baseline). End-tidal forcing was employed for the control of blood gases. To examine if INDO affected ICA vasomotion in a cyclooxygenase inhibition independent manner, a subset of subjects (n=5) were tested before and 45-minutes following oral Ketorolac (0.25mg/kg). During pre-drug testing in the INDO trial, the ICA dilated during hypercapnia at +6mmHg (4.72±0.45 vs. 4.95±0.51mm; P<0.001) and +9mmHg (4.72±0.45 vs. 5.12±0.47mm; P<0.001), and constricted during hypocapnia at -6mmHg (4.95±0.33 vs. 4.88±0.27mm; P<0.05) and -9mmHg (4.95±0.33 vs. 4.82±0.27mm; P<0.001). Following INDO, dilation of the ICA was still observed at +6mmHg (4.50±0.54 vs. 4.57±0.52mm; P<0.05) and +9mmHg (4.50±0.54 vs. 4.61±0.50mm; P<0.01); however, INDO reduced the vasomotor responsiveness by 67±28% (0.045±0.015 vs. 0.015±0.012mm ⋅ mmHgPETCO2-1). In the Ketorolac condition, there was no effect of the drug on the vasomotor response to hyper or hypocapnia. We conclude that: 1) changes in PETCO2 mediate vasomotion of the ICA, 2) inhibition of non-selective prostaglandin synthesis via INDO markedly reduces the vasomotor response to changes in PETCO2; and 3) INDO may be acting via a mechanism(s) independent of cyclooxygenase inhibition to reduce CO2 mediated vasomotion.    ! iii!Preface  Chapter 1. Aspects from Chapter 1 (see “1.4 Measurement of cerebral blood flow”) have been published elsewhere: Ainslie PN., Hoiland RL. (2014) Transcranial Doppler Ultrasound: Valid, Invalid, or Both? Journal of Applied Physiology. 117(10), 1081-1083. Both the editorial first draft and figure 1 were created in cooperation with Prof. Ainslie. I completed the revisions following the editor’s comments in cooperation with Prof. Ainslie. ©Permission was not required for reproduction of figures and text upon request. I wrote Chapter 1, and received extensive feedback from Prof. Ainslie through several editing processes.  Chapter 2. Chapter 2 will be submitted to J Physiol for publication. Prof. Ainslie and myself planned the experiment. Data collection was completed with the assistance of Mike Tymko, Anthony Bain, and Kevin Wildfong. Dr. Brad Monteleone provided useful discussions for the selection of our drug interventions. This study was completed at The University of British Columbia – Okanagan Campus. I completed all data analysis and wrote the manuscript. Prof. Ainslie provided extensive feedback and critically reviewed the manuscript for appropriate data interpretation and content. All co-authors edited the manuscript. This study received ethical approval from The University of British Columbia Clinical Research Ethics Board (ID: H11-02657).  Chapter 3. I wrote chapter 3 and received extensive feedback from Prof. Ainslie prior to finalization.             ! iv!Table of Contents  Abstract ................................................................................................................................... ii Preface .................................................................................................................................... iii Table of Contents .................................................................................................................. iv List of Tables ........................................................................................................................ viii List of Figures ........................................................................................................................ ix List of Abbreviations .............................................................................................................. x Acknowledgements ............................................................................................................... xii Dedication ............................................................................................................................ xiii Chapter 1: Introduction ......................................................................................................... 1 1.1. An overview of cerebrovascular regulation………….……….….……………1 1.1.1. CBF regulation by arterial blood gases ................................................ 2 1.1.2. CBF regulation by cerebral metabolic processes..................................3 1.1.3. CBF regulation by cerebral perfusion pressure .................................... 5 1.1.4. CBF regulation by the autonomic nervous system ............................... 7 1.2. Cerebrovascular regulation by arterial carbon dioxide………….…………..7 1.2.1. Response characteristics ....................................................................... 8 1.2.2. Factors influencing cerebrovascular CO2 reactivity ........................... 11 1.2.3. The potential cellular mechanisms mediating cerebrovascular  CO2 reactivity......................................................................................12 1.2.3.1. Potassium channels.................................................................13 1.2.3.2. Adenosine...............................................................................14 1.2.3.3. Nitric oxide.............................................................................15 1.2.3.4. Epoxyeicosatrienoic acids......................................................15 1.3. The prostaglandin pathway .............................................................................. 16 1.3.1. Prostaglandins and animal cerebral vasomotor tone: In Vitro ........... 17 1.3.2. Prostaglandins, animal CBF and CO2 reactivity: In Vivo .................. 18 1.3.3. Prostaglandins and human cerebral vasomotor tone: In Vitro........... 19 1.3.4. Prostaglandins and human CBF: In Vivo ........................................... 21 1.3.5. Prostaglandins and human cerebrovascular CO2 reactivity:  In Vivo . .............................................................................................. 22 ! v!1.3.6. Prostaglandin pathway: cyclic adenosine monophosphate production........................................................................................... 26 1.3.7. Prostaglandin pathway: reactive oxygen species production ............. 28 1.3.8. Prostaglandin pathway: interaction with nitric oxide......................... 29 1.4 Measurement of cerebral blood flow ................................................................ 30   1.4.1. Transcranial Doppler ultrasound.......................................................... 31 1.4.1.1. Doppler physics ..................................................................... 33 1.4.1.2. Vessel location and insonation .............................................. 33                                     1.4.1.3. The trans-temporal approach ................................................. 34                                     1.4.1.4. Quantification of cerebral blood velocity from transcranial Doppler ultrasound ................................................................. 34                         1.4.2. Duplex ultrasound of extra-cranial cerebral blood vessels .................. 34                                     1.4.2.1. Location and insonation of the internal carotid artery .......... 35                                     1.4.2.2. Quantification of CBF from duplex ultrasound .................... 36                                     1.4.2.3. Analysis software .................................................................. 38                          1.4.3. Variability in assessing cerebrovascular CO2 reactivity ..................... 40                                     1.4.3.1. Measurement variability ........................................................ 41             1.5. Experimental models to assess cerebrovascular CO2 reactivity ................... 42                          1.5.1. Changes in fractional inspired CO2 ..................................................... 43                          1.5.2. The rebreathe method .......................................................................... 43                          1.5.3. End-tidal forcing .................................................................................. 44             1.6. Purpose and hypotheses .................................................................................... 44 Chapter 2: Carbon dioxide mediated vasomotion of extra-cranial cerebral arteries:          a role for prostaglandins? ................................................................................ 46  2.1. Overview ............................................................................................................ 46              2.2. Background . ..................................................................................................... 47              2.3. Methods ............................................................................................................. 49                           2.3.1. Subjects .............................................................................................. 49                           2.3.2. Experimental protocol ........................................................................ 49                                      2.3.2.1. Study 1 .................................................................................. 49                                      2.3.2.2. Study 2 .................................................................................. 50 ! vi!                          2.3.3. Experimental measures ....................................................................... 51                                      2.3.3.1. Cardiorespiratory measures .................................................. 51                                      2.3.3.2. Cerebrovascular measures .................................................... 51                           2.3.4. Statistical analyses .............................................................................. 52              2.4. Results  ............................................................................................................... 53                           2.4.1. Resting cerebral blood flow and cardiorespiratory variables ............. 53                           2.4.2. Cerebrovascular CO2 reactivity (Table 2.0) ....................................... 53                           2.4.3. Vasomotor response of the internal carotid artery to CO2 ................. 56                                        2.4.4. Effect of Ketorolac on CO2 mediated responses (Table 2.1) ............. 58              2.5. Discussion .......................................................................................................... 59                           2.5.1. Cerebrovascular responses to CO2 ..................................................... 59                           2.5.2. Comparisons between studies ............................................................ 60                           2.5.3. Cyclooxygenase inhibition ................................................................. 61                           2.5.4. Pharmacological interventions ........................................................... 62                           2.5.5. Implications  ....................................................................................... 63              2.6. Synopsis  ............................................................................................................ 64               2.7. Author contributions ....................................................................................... 65              2.8. Funding ............................................................................................................. 65              2.9. Special Recognition .......................................................................................... 65 Chapter 3: Conclusion ......................................................................................................... 66              3.1. Indomethacin induced impairments of cerebrovascular reactivity ............. 66              3.2. Cerebrovascular CO2 reactivity: Implications for disease ........................... 67                           3.2.1. Sleep apnea ......................................................................................... 67                           3.2.2. Prediction of mortality ........................................................................ 68              3.3. Methodological limitations .............................................................................. 69                           3.3.1. Pharmacological inhibition of cyclooxygenase .................................. 70                           3.3.2. End-tidal versus arterial CO2 .............................................................. 70              3.4. Future studies ................................................................................................... 70                            3.4.1. Determining the role of cyclic adenosine monophosphate                                       inhibition on cerebrovascular vasomotor responses in humans ........ 72    3.4.2. The assessment of cerebral endothelial function ............................... 72 ! vii!                           3.4.3. Vasomotor responsiveness of the internal carotid artery with                                       aging .................................................................................................. 73                            3.4.4. Other regulatory factors in the cerebral vasomotor response                                       to CO2 ............................................................................................... 73 Bibliography ......................................................................................................................... 74  ! viii!List of Tables  Table 1.0.  Cyclooxygenase inhibitors used to investigate the role of prostaglandins  on cerebral blood flow .................................................................................... 22 Table 1.1.  The effect of cyclooxygenase inhibitors on resting cerebral blood flow  and cerebrovascular CO2 reactivity ................................................................ 25 Table 2.0.  Cerebrovascular and blood pressure responses to CO2 before and  following Indomethacin or placebo ................................................................ 54 Table 2.1.  Cerebrovascular and blood pressure responses to CO2 before and  Following Ketorolac ....................................................................................... 59   ! ix!List of Figures  Figure 1.0.   The steady state cerebral pressure-flow relationship in humans ...................... 6 Figure 1.1.   Carbon dioxide mediated cerebral vasomotion occurs throughout the  entire cerebral vascular tree .............................................................................. 8 Figure 1.2.       Regional differences in cerebrovascular reactivity to carbon dioxide ........... 10 Figure 1.3.   The putative mechanisms governing cerebrovascular smooth muscle  cell relaxation during CO2 perturbations ........................................................ 13 Figure 1.4.      The arachidonic acid pathway ........................................................................ 17 Figure 1.5.      Putative pathways for prostanoid mediated signal transduction ..................... 20 Figure 1.6.   Absolute cerebrovascular CO2 reactivity during hyperoxic  hypercapnia pre- and post 1.45mg/kg indomethacin in humans .................... 23 Figure 1.7.   The impact of middle cerebral artery vasomotion on the discrepancy  between volumetric and velocity indices of cerebral blood flow ................... 32 Figure 1.8.      Example duplex ultrasound image of the internal carotid artery .................... 38 Figure 1.9.   Example analysis software data output...........................................................40 Figure 1.10.  Bland-Altman plots of duplex ultrasound measurements of volumetric  blood flow through the internal carotid artery ................................................ 41 Figure 1.11.  Bland-Altman plots of transcranial Doppler ultrasound measures of  blood velocity through the middle cerebral artery ......................................... 42 Figure 2.0.       Volumetric flow and velocity cerebrovascular reactivity to hypercapnia ..... 55 Figure 2.1.       Volumetric flow and velocity cerebrovascular reactivity to hypocapnia ...... 56 Figure 2.2.   The vasomotor response to hypercapnia.........................................................57 Figure 2.3.   The vasomotor response to hypocapnia..........................................................58   ! x!List of Abbreviations  Abbreviation   Definition 20-HETE    20- hydroxyeicosatetraenoic acid AA     Arachidonic acid ACA     Anterior cerebral artery ATP     Adenosine triphosphate BA     Basilar artery BOLD     Blood oxygen level dependent Ca2+     Calcium cAMP     Cyclic adenosine monophosphate CBF     Cerebral blood flow cGMP     Cyclic guanosine monophosphate CO2     Carbon dioxide CoV     Coefficient of variation COX     Cyclooxygenase CSA    Cross-sectional area EET     Epoxyeicosatrienoic acid FiCO2     Fractional inspired carbon dioxide HR     Heart rate ICA     Internal carotid artery ICAv     Internal carotid artery blood velocity INDO     Indomethacin K+     Potassium KATP channels    Adenosine triphosphate sensitive potassium channels MAP     Mean arterial pressure  MCA     Middle cerebral artery MCAv    Middle cerebral artery blood velocity MLCK    Myosin light chain kinase MRI     Magnetic resonance imaging nNOS     Neuronal nitric oxide synthase NO     Nitric oxide ! xi!NOS     Nitric oxide synthase NVC     Neurovascular coupling O2     Oxygen PaCO2     Partial pressure of arterial carbon dioxide PaO2     Partial pressure of arterial oxygen PCA     Posterior cerebral artery PCAv     Posterior cerebral artery blood velocity PETCO2    Partial pressure of end-tidal carbon dioxide PETO2     Partial pressure of end-tidal oxygen PG     Prostaglandin PGD2     Prostaglandin D2 PGE2     Prostaglandin E2 PGF2α     Prostaglandin F2α PGH2     Prostaglandin H2 PGI2     Prostacyclin PGG2     Prostaglandin G2 PKA     Cyclic adenosine monophosphate dependent protein kinase PKG     Cyclic guanosine monophosphate dependent protein kinase PLA2     Phospholipase A2 QICA     Internal carotid artery blood flow ROI     Region of interest ROS     Reactive Oxygen Species TCD     Transcranial Doppler ultrasound VA     Vertebral artery Vmax     Maximum blood velocity            ! xii!Acknowledgments  First, I would like to thank my supervisor Prof. Ainslie for his continued support that began in the third year of my undergraduate degree through to the completion of my Masters degree. You have provided me with some amazing opportunities - working in the EV-K2-CNR Pyramid Research Laboratory and studying elite breath-hold divers in Croatia - that I did not know were possible in an academic setting. Through your teaching not only have I developed academically, but I have above all else learned the importance of pursuing your passions and the importance of work / lifestyle balance.  Thank you to my committee, Drs. Neil Eves and Glen Foster for their continued help throughout my Masters degree. I must also thank Dr. Trevor Day who has been an important mentor throughout my Masters degree.  I must thank Lauren Ray for introducing me to the Ainslie laboratory in my third year. Without that introduction this work would have never came to fruition.  To Brooke Madsen, thank you for your continued patience as I worked with a sense of urgency that so often disrupted our plans. Your love and support have been vital to my sanity throughout this experience, and I could not have done it without you.  I am forever indebted to my loving parents Ivar and Marcelle Hoiland, for helping shape me into the individual I am and preparing me for the transitions that occur throughout life. Without your guidance and support I would not have been able to pursue an education. I am forever grateful.           ! xiii!         For Ivar and Marcelle    Here lies the product of 23 years hard work      ! 1!Chapter 1: Introduction   1.1. An overview of cerebrovascular regulation.   The human brain is distinctly different from other organs in that it possesses a very high metabolic demand with limited intra-cellular energy stores, thus necessitating precisely controlled blood flow to match metabolic demand. A disproportionately high metabolic demand relative to its size (~2% of total body weight) necessitates the direction of approximately 15% of resting cardiac output to the brain. Ultimately, the brain is responsible for ~20% of resting metabolism (Kety & Schmidt, 1948a). As a result, maintenance of optimal oxygen (O2) and nutrient supply is paramount in maintaining neuronal function and consciousness (Lennox et al., 1935; Van Lieshout et al., 2003). Complete cessation of cerebral blood flow (CBF) results in syncope in as little as four seconds, which can quickly progresses to more serious outcomes such as seizure, permanent neurological damage, and eventually death if cerebral perfusion is not restored within a few minutes (Smith et al., 2011). Precise control of CBF involves the interplay of key and often overlapping or redundant regulatory mechanisms. The four primary regulators of CBF, which encompass their own respectively complex regulation, are arterial blood gases [i.e., O2 and carbon dioxide(CO2); Kety & Schmidt, 1948b; Willie et al., 2012; Wolff & Lennox, 1930], cerebral metabolism [i.e., neurovascular coupling (NVC); Roy & Sherrington, 1890; Attwell et al., 2010], cerebral perfusion pressure (i.e., cerebral autoregulation; Lucas et al., 2010; Tzeng & Ainslie, 2014; Numan et al., 2014), and to a lesser extent regulation by the autonomic nervous system (e.g., Umeyama et al., 1995).   The brain’s high blood supply is delivered through two bilateral pairs of large arteries: two internal carotid arteries (ICA), and two vertebral arteries (VA). The VA’s run cephalad off of the subclavian artery, through processes in the vertebral column starting at the 6th cervical vertebrae, and ramify to form the basilar artery (BA) at the base of the brain. The ICA’s run cephalad from the common carotid bifurcation to the base of the brain, where they along with communicating arteries, the BA, and the posterior cerebral arteries (PCA) ramify into an anastomotic ring, called the Circle of Willis. This unique anastomotic ring was first reported ! 2!to exist in 1664 in the doctoral works of Sir Thomas Willis titled: Cerebri Anatome (Willis, 1664).  The circle of Willis gives rise to three pairs of large cerebral arteries: The anterior cerebral arteries (ACA), middle cerebral arteries (MCA), and the aforementioned PCA’s. These large cerebral arteries run outward towards the surface of the cerebral cortex, exiting the brain parenchyma where they then enter into the pia mater. At this point, they are referred to as pial vessels, which were previously thought to be the exclusive regulators of cerebrovascular resistance (Wolff & Lennox, 1930). However, it is now known that while pial vessels may be the primary regulator of cerebrovascular tone, vasomotion of large cerebral arteries also plays a role in mediating cerebrovascular resistance in both animals (Heistad et al., 1978; Faraci & Heistad, 1990) and humans (Willie et al., 2014; Lewis et al., 2015). These pial vessels, which transition into penetrating arterioles, dive down into the brain parynchema through what is known as the ‘Virchow Robin’ space. Penetrating arterioles become encapsulated in astrocytic end-feet and pericytes, and transition into parynchemal arterioles. The regulation of CBF by arterial blood gases, metabolism, cerebral perfusion pressure, and the autonomic nervous system will be discussed next.   1.1.1. CBF regulation by arterial blood gases.   The cerebral vasculature is highly sensitive to changes in the partial pressure of arterial carbon dioxide, and to a lesser extent, oxygen (PaCO2 & PaO2, respectively). Changes in PaCO2 bidirectionally influence CBF as elevations in PaCO2 (hypercapnia) cause an increase in CBF and reductions in PaCO2 (hypocapnia) cause a decrease in CBF (Ainslie & Duffin, 2009; Willie et al., 2012). While PaO2 is not as potent a regulator of CBF as PaCO2, a marked reduction in PaO2 (i.e., to below 55mmHg) results in a compensatory increase in CBF to maintain cerebral O2 delivery during both normobaric (see: Figure 1 in: Ainslie et al., 2014) and hypobaric (see: Figure 2 in: Ainslie & Subudhi, 2014) hypoxia. Hyperoxia (elevated PaO2) can have a small constrictive influence on CBF, that is largely mediated via hyperventilation (Willie et al., 2012; Smith et al., 2012). With respect to the influence of blood gases on blood flow regulation, the cerebral vasculature is distinctly different from that ! 3!of the peripheral circulation (Lennox & Gibbs, 1932; Ainslie et al., 2005) where there is a limited influence of arterial blood gases. This impact of changes in arterial blood gases on cerebrovascular regulation was observed nearly a century ago in cats (Wolff & Lennox, 1930), soon thereafter investigated in humans (Lennox & Gibbs, 1932), and then comprehensively characterized several decades later (Shapiro et al., 1966). Cerebrovascular reactivity to changes in PaCO2 is important in maintaining constancy of central pH and hence breathing stability (Ainslie & Duffin, 2009; Xie et al., 2009; Fan et al., 2010). For example, impaired (reduced) cerebrovascular reactivity to PaCO2 is implicated in central sleep apnea, in both patients with congestive heart failure (Xie et al., 2005; Javaheri & Dempsey, 2013) and in otherwise healthy volunteers at high altitude (Burgess et al., 2014). Moreover, impaired cerebrovascular reactivity to PaCO2 is associated with an increased risk for stroke (Markus & Cullinane, 2001) and all cause mortality (Portegies et al., 2014). Therefore, in addition to maintaining O2 delivery during hypoxia, it is obvious that cerebrovascular reactivity to blood gases is essential in maintaining homeostasic function in several capacities.  1.1.2. CBF regulation by cerebral metabolic processes.   Increases in neural activity induce metabolic demand for O2 and glucose, and are coupled to increases in CBF (Roy & Sherrington, 1890; Attwell et al., 2010; Willie et al., 2011b). Much of the data informing scientific knowledge of the coupling between cerebral metabolism and CBF have been derived from various gas perturbations and pharmacological blockades in vitro using brain tissue slices, while a relatively smaller number of studies have sought to investigate the regulation of metabolic and CBF coupling on a global scale (i.e., total brain). The phenomena of metabolism and flow coupling, both in tissue bath preparations and in vivo is termed - and will henceforth be referred to as - neurovascular coupling (NVC). This energy consumption of the brain, which drives NVC, stems primarily from maintenance of ionic gradients (related to action potentials, resting potentials, etc.), with metabolic demand increasing in concert with increased neuronal firing (Attwell & Laughlin, 2001).  ! 4!Neural activity causes an immediate increase in dendritic glycolysis with the metabolic response in astrocytes delayed slightly (Kasischke et al., 2004). The immediate increase in dendritic activity is driven primarily by oxidative metabolism, and seemingly depletes O2 stores, which results in the astrocytic response being primarily dependent on anaerobic glycolysis (Kasischke et al., 2004). The NVC occurs in mere seconds, and acts to mitigate and effectively eliminate the immediate drop in tissue O2 associated with the onset of neuronal activity (Vanzetta & Grinvald, 1999; Offenhauser et al., 2005). Larger increases in neural activity elicit larger reductions in tissue O2 and concomitant compensatory increases in CBF, while impairments in NVC potentiate the drop in tissue O2 (Offenhauser et al., 2005). Astrocytes seem to be the primary regulator of the NVC response, with the ability to modulate pre-synaptic, post-synaptic, and direct vascular signaling due to their involvement in what is known as the ‘tripartite synapse’ and their direct apposition to microvasculature (Perea et al., 2009; Kowiański et al., 2013).  The current leading theory is that NVC acts as a feed forward mechanism, as synaptic release of glutamate and post-synaptic metabotropic glutamate receptor binding (i.e., simulated neural activity) results in an increase in intracellular astrocytic Ca2+ leading to adjacent microvascular dilation (Attwell et al., 2010). At a fundamental level, the signal mediating both neural activity and NVC is one in the same (glutamate receptor binding), highlighting the efficacy of the system mediating NVC. Several primary factors have been proffered to regulate NVC including: 1) prostaglandins (PG), 2) adenosine, 3) nitric oxide (NO), and 4) the prevailing level of O2 tension [reviewed in: (Attwell et al., 2010)]. Briefly, increased intracellular Ca2+, from metabatropic glutamate receptor binding or photolysis of caged Ca2+, signals the conversion of membrane phospholipids into arachidonic acid and subsequently prostaglandin E2 (PGE2). This PGE2 moves out of the astrocyte and into the perivascular space where it binds to prostaglandin EP receptors on adjacent vascular smooth muscle, resulting in smooth muscle relaxation (Gordon et al., 2008). In instances where astrocytic O2 levels are reduced (i.e., increased glycolysis), adenosine triphosphate (ATP) production decreases increasing the intracellular astrocytic adenosine concentration. Consequently, adenosine is released to the perivascular space where it binds to adenosine A2A receptors on adjacent smooth muscle cells, with the consequent signal transduction inhibiting Ca2+ ! 5!channel mediated increases in intracellular vascular smooth muscle Ca2+, thus inhibiting constriction (Gordon et al., 2008). Lastly, the prevailing level of O2 has been demonstrated to dictate the directionality that these mechanisms conform to in response to neuronal activation. In the brain-slice preparation used by Gordon et al., 2008, hyperoxia resulted in neuronal activation mediated microvascular vasoconstrictions whereas physiological levels of O2 (i.e., 20% O2 in tissue bath) ‘switched’ the NVC response to mediate microvascular vasodilation (Gordon et al., 2008)  While the mechanistic underpinnings of NVC are primarily determined in vitro, NVC has been quantified in vivo in humans using several measurement techniques including blood oxygen level dependent (BOLD) magnetic resonance imaging (MRI; Bruhn et al., 2001), arterial spin tagging MRI (St Lawrence et al., 2003), and transcranial Doppler ultrasound (TCD; Willie et al., 2011). Little work has been done to determine the specific cellular pathways contributing to NVC in living humans. Therefore, the in vivo origins and regulation of NVC remains poorly understood.   1.1.3. CBF regulation by cerebral perfusion pressure.   Early study of the cerebrovascular response to changes in blood pressure dates back nearly one hundred years. In the late 1930’s Mogens Fog published two papers that reported decreases in blood pressure lead to increases in pial arteriolar diameter (Fog, 1937), while increases in blood pressure resulted in decreases in pial arteriolar diameter (Fog, 1939). These changes occur due to active vasomotion (smooth muscle cell contraction / relaxation), with each response taking ~30-60 seconds to initiate. This counteractive effect of pial arterial vasomotion in response to changes in blood pressure was later misconstrued to maintain constancy of CBF during changes in cerebral perfusion pressure and coined “cerebral autoregulation” by Niels Lassen in 1959 (Lassen, 1959). This theory is still to this day misconstrued and published in high impact journals (i.e., Meng & Gelb, 2015) despite our much improved understanding of cerebral pressure flow relationships (Tzeng & Ainslie, 2014). In actuality, CBF is dependent on cerebral perfusion pressure in a primarily passive manner (Figure 1.0). While changes in pial arteriolar caliber may act to protect the cerebral ! 6!microvasculature from deleterious changes in intravascular pressure, CBF changes in concert with cerebral perfusion pressure in both steady state (Lucas et al., 2010; Numan et al., 2014) and dynamic (Tzeng et al., 2010; Tan, 2012) instances. The presence of segmentally specific regulation of vasomotor tone by blood pressure, at least in animals, provides evidence for this teleological perspective of protecting the microvasculature. For example, during increases in blood pressure, the larger pial vessels constrict progressively in a concomitant fashion, while the smallest pial vessels dilate (Kontos et al., 1978). Conversely, larger pial vessels dilate initially upon a reduction in blood pressure, while the smaller arterioles begin to dilate at a lower pressure threshold (Kontos et al., 1978). This highlights that there is not one site of cerebrovascular resistance governing the relationship between pressure and flow, and that cerebrovascular regulation during blood pressure perturbations is a highly complex and integrative process. Overall, the cerebrovasculature seems to be more sensitive to decreases than increases in CBF (Tzeng et al., 2010; Numan et al., 2014; Figure 1.0), which may involve elevations in cerebral specific sympathetic nervous activity (SNA) during increases in blood pressure (Cassaglia et al., 2008).    Figure 1.0. The steady state cerebral pressure-flow relationship in humans. Changes in mean arterial pressure (MAP) lead to pressure passive changes in cerebral blood flow (CBF) in humans. The above figure was synthesized from 40 peer-reviewed articles with a total of 49 separate experiments. Thin lines represent the mean cerebral pressure-flow relationship of individual studies while the weighted mean of all studies is depicted by the thick black lines. These data collectively depict that CBF changes in concert with MAP, while the cerebrovasculature is less sensitive to increases in MAP (Numan et al., 2014). From: Numan et al., Med Eng & Phys, 2014 with ©permission.  ! 7!1.1.4. CBF regulation by the autonomic nervous system.   The human cerebral circulation is highly innervated with perivascular nerves, with the highest density of nerves located in the posterior arteries (Bleys et al., 1996). Previous study has shown that ganglionic blockade at rest in healthy individuals has no effect on MCA blood velocity (MCAv; Ide et al., 2000; Zhang & Levine, 2007) or extra-cranial vessel (ICA & VA) diameter as assessed with MRI (Kang et al., 2010). However, these findings are not universal as CBF has been reported to increase with ganglionic blockade when measured with single photon emission computed tomography (Umeyama et al., 1995). Thus, while the role of the autonomic nervous system in regulating CBF remains controversial, it is likely quite a negligible factor at rest. To the contrary, there is some evidence that the autonomic nervous system, specifically, the sympathetic nervous system is integral in regulating CBF during surges in blood pressure. As mentioned previously (see section “1.1.3 CBF regulation by cerebral perfusion pressure”), increases in blood pressure result in less of a CBF response than do decreases in blood pressure. It is thought that while increases in systemic blood pressure activate the baroreflex and actively reduce sympathetic output to the periphery, that there is a cerebral specific increase in SNA (Cassaglia et al., 2008), although this has only been demonstrated in animals.  1.2. Cerebrovascular regulation by arterial carbon dioxide.   Of the aforementioned regulators of CBF (metabolism, perfusion pressure, arterial blood gases and autonomic nervous activity), PaCO2 is seemingly the most potent regulator of cerebrovascular tone. The cerebrovascular response to CO2 is characterized as the unit (i.e., mL/min), or percent, change in CBF per unit change in either PETCO2 or PaCO2 – termed cerebrovascular CO2 reactivity. Changes in PaCO2 mediate alterations in CBF locally via changes in extravascular pH (Kontos et al., 1977a), but not intraluminal CO2 (Kontos et al., 1977b) or pH (Lambertsen et al., 1961; Harper & Bell, 1963). For example, in cats, pretreatment with intravenous bicarbonate to maintain normal pH nearly abolishes pial vessel dilation in response to hypercapnia (Kontos et al., 1977b) highlighting the dependency of this response on pH. Therefore, movement of CO2 through the vessel wall and consequent ! 8!alteration of extravascular pH is seemingly necessary to alter vasomotor tone. While previously a contentious matter (Serrador et al., 2000; Giller, 2003), evidence continues to emerge supporting the theory that CO2 is vasoactive throughout the entire cerebral vascular tree (see Figure 1.1) implicating both large cerebral arteries through to cerebral arterioles in the regulation of cerebrovascular resistance during PaCO2 perturbations (Wolff & Lennox, 1930; Willie et al., 2012; Verbree et al., 2014; Coverdale et al., 2014, 2015). Previous animal data corroborates this recent paradigm shift in human cerebrovascular regulation (Heistad et al., 1978; Faraci & Heistad, 1990).    Figure 1.1. Carbon dioxide mediated cerebral vasomotion occurs throughout the entire cerebral vascular tree. A. Classic data from Wolff & Lennox (Wolff & Lennox, 1930) depicting hypercapnic vasodilation of pial arterioles in anesthetized cats. Two vessels can be seen; the one on the left is an image during craniotomy in the resting state, while the one on the right is during hypercapnia. The difference in diameter can be visualized by the addition of the dotted line to the diameter of the vessel in the right image. These data, along with extrapolation from others (Serrador et al., 2000), led to the belief that CBF regulation during changes in PaCO2 was mediated solely by pial arteries / arterioles. Adapted from Wolff & Lennox, 1930 - ©permission not required upon request. B. Vasomotion of the MCA as assessed by high resolution (3T) MRI in humans during changes in end-tidal carbon dioxide (PETCO2).  Reproduced from Coverdale et al., 2014 - ©permission not required upon request. C. Data depicting vasomotion of the internal carotid artery (ICA) in humans throughout a wide range of arterial PCO2 (PaCO2). It is now, therefore, established that vasomotion of large intra- and extra- cranial cerebral arteries occurs in response to changes in PaCO2 along with changes at the pial artery and arteriolar level. Reproduced from Willie et al., 2012 - ©permission received.    1.2.1. Response characteristics.   Cerebrovascular CO2 reactivity is a relatively linear response during both hypo and hypercapnia (Skow et al., 2013), with the magnitude of reactivity during hypercapnia near double that during hypocapnia (i.e., hypercapnia ~ Δ4% CBF/mmHgPaCO2 vs. hypocapnia ~ Δ2%CBF/mmHgPaCO2; Ainslie & Duffin, 2009; Willie et al., 2012; Figure 1.2-B). There are ! 9!regional differences in the magnitude of reactivity between the anterior and posterior circulation, and between grey and white matter. Specifically, grey matter possesses an approximate 3-fold greater CO2 reactivity compared to white matter (Ramsay et al., 1993), likely due do to lower vascularization.   When assessed using duplex ultrasound, absolute CBF reactivity of the anterior circulation through the ICA is approximately double that of posterior reactivity through the VA during hypocapnia (Willie et al., 2012) with the difference being even greater during hypercapnia (Willie et al., 2012; Hoiland et al., 2015). Differences in absolute MCAv and posterior cerebral artery blood velocity (PCAv) reactivity are not apparent during end-tidal forcing through an extreme range of changes in PaCO2 (i.e., 15-65mmHg; Willie et al., 2012) or during hyperventilation and changes in fractional inspired CO2 (Ogawa et al., 1988). However, apparent differences in hypercapnic CBF reactivity between anterior and posterior circulation (ICA vs. VA) are corroborated by MCAv and PCAv reactivity measures during hyperoxic rebreathing (Skow et al., 2013). Differences between Skow et al., 2013 and Willie et al., 2012 may be due to differences in statistical power (i.e., sample size), while Ogawa et al., 1988 utilized non-linear analysis parameters as opposed to the more commonly used linear analysis approach. Thus it is fair to conclude the anterior circulation has a higher absolute reactivity than that of the posterior circulation, at least as observed using duplex ultrasound.  ! 10! Figure 1.2. Regional differences in cerebrovascular reactivity to carbon dioxide. The above figure outlines both intra- and extra-cranial reactivity to CO2. A. Individual vessel relative reactivity to changes is PaCO2 ranging from ~15mmHg to 65mmHg. B. Individual vessel relative reactivity calculated with linear regression in the hypercapnic range and hypocapnic range. Relative hypercapnic reactivity was not different between vessels, while the VA showed selectively higher reactivity than all other vessels during hypocapnia. * significantly different from all other vessels, P<0.05. Reproduced from Willie et al., 2012 - ©permission received.   In relative terms (i.e., %ΔCBF ⋅ mmHgPaCO2-1), the majority of data indicate that both ICA and VA flow/velocity reactivity and MCAv and PCAv reactivity do not differ during hypercapnia (Hauge et al., 1980; Willie et al., 2012; Skow et al., 2013; Hoiland et al., 2015), with very minimal evidence to the contrary (Sato et al., 2012). Similarly, both MCAv and PCAv and ICA flow reactivity do not differ during hypocapnia; however, VA flow reactivity has been reported as both greater than (Hauge et al., 1980; Willie et al., 2012) and similar (Sato et al., 2012) to that of the ICA during hypocapnia. Between-study differences likely relate to differing experimental paradigms, manipulations of blood gases, and measurement technique standardization and analysis. The specifics of measuring CBF with ultrasound and different methods for blood gas manipulation are discussed in Section 1.4. Measurement of cerebral blood flow; and Section 1.5. Experimental methods to assess cerebrovascular CO2 reactivity.  ! 11!1.2.2. Factors influencing cerebrovascular CO2 reactivity.  Cerebrovascular CO2 reactivity is subject to between test and between subject variability and can be altered by a myriad of potential concurrent physiological effectors. Consideration of potential interactions with other primary regulators of CBF (hypoxia, neuronal activity, blood pressure, and autonomic inputs) is pivotal in fully understanding CO2 induced vasomotion. It has been previously demonstrated that alterations in PaO2 will interact with PaCO2 perturbations to collectively dictate cerebrovascular reactivity. Specifically, hypoxia will elevate CBF across a wide range of PaCO2 in both the hypo and hypercapnic CO2 range (Ainslie & Poulin, 2004; Mardimae et al., 2012). In instances of severe hypoxia (i.e., PaO2≈40mmHg), CBF remains above baseline despite pronounced hypocapnia (Mardimae et al., 2012).  Elevations in PaCO2 result in concomitant increases in blood pressure due to increased SNA (Ainslie et al., 2005). When the elevations in blood pressure are sustained, they are generally related to the magnitude of the flow response to hypercapnia (Willie et al., 2012; Regan et al., 2014). Thus, it is important to consider the influence of systemic blood pressure in the cerebrovascular response to CO2. A seminal study conducted in 1965 by Harper & Glass, showed that progressive decreases in mean arterial pressure (MAP) up to a reduction of 66% below baseline, causes marked reductions in cerebrovascular CO2 reactivity to hypo and hypercapnia (Harper & Glass, 1965). It is unclear if progressive elevations in blood pressure also affect the cerebrovascular response to changes in PaCO2.   Lastly, chemoreflex activation via acidosis increases SNA (Steinback et al., 2009, 2010b, 2010a) and may, therefore, be implicated in modulating cerebrovascular CO2 reactivity. Increasing SNA through various techniques does not alter reactivity in either the hyper and hypocapnic range (LeMarbre et al., 2003; Ainslie et al., 2005); however, pharmacological blockade of SNA has been reported to reduce both hypocapnic (Peebles et al., 2012) and hypercapnic (Przybyłowski et al., 2003) reactivity.  A reduction in hypercapnic reactivity via pharmacological SNA blockade is likely mediated by an abolished pressor response (Przybyłowski et al., 2003) which, as mentioned, normally contributes to the flow response ! 12!(Willie et al., 2012; Regan et al., 2014). There is, however, data indicating that the pressor response is augmented by ganglionic blockade (Jordan et al., 2000). This relationship between SNA and CO2, and cerebrovascular CO2 reactivity has yet to be investigated using volumetric flow measurement techniques. Collectively, it is apparent that while cerebrovascular CO2 reactivity is often viewed as an independent effector of CBF, it is in fact an integrative response that interacts with many other physiological variables.   1.2.3. The potential cellular mechanisms mediating cerebrovascular CO2 reactivity.   Regulation of CBF at the cellular level in response to varying stimuli is a multifaceted and complex process. See Figure 1.3 for a generalized overview of cerebrovascular smooth muscle cell regulation by CO2. Overall, potassium channels and vasoactive factors such as adenosine, NO, epoxyeicosatrienoic acids (EET’s), and PG’s all potentially mediate cerebrovascular vasomotion in response to alterations in PaCO2. Each of these factors will be briefly overviewed below, while the role of PG’s will be reviewed in detail.     ! 13! Figure 1.3. The Putative mechanisms governing cerebrovascular smooth muscle cell relaxation during CO2 perturbations. Changes in potassium channel conductance and alterations in vascular smooth muscle cell calcium sensitivity mediated via cyclic nucleotides govern vascular tone. Increased potassium channel conductance results in potassium efflux from, and a hyperpolarization of, smooth muscle cells leading to relaxation, while reductions in conductance lead to contraction (Faraci & Sobey, 1998). Calcium sensitivity is regulated primarily by cyclic nucleotide activity, cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP). For example, increases in cAMP activate cAMP dependent protein kinase (PKA), which inhibits smooth muscle cell contraction (Kerrick & Hoar, 1981) through phosphorylation (and deactivation) of myosin light chain kinase (MLCK; Adelstein & Conti, 1978). Moreover, cyclic nucleotides increase conductance through potassium channels (Song & Simard, 1995). Prostaglandins (PGE2 & PGI2), epoxyeicosatrienoic acids (EET’s), nitric oxide (NO), and adenosine are all capable of modulating cyclic nucleotide activity and / or potassium channel conductance. In the above figure a (+) symbol represent activation of a downstream signal, whereas (-) represents deactivation. AA, Arachidonic Acid; PKG, cGMP dependent protein kinase; PLA2, Phospholipase A2.   1.2.3.1. Potassium channels.   Potassium (K+) channels are present in cerebrovascular smooth muscle cells and their impact on membrane potential is an important regulator of vascular tone (Faraci & Sobey, 1998). Activation of K+ channels increases K+ conductance resulting in efflux of K+ from vascular smooth muscle cells and consequent hyperpolarization and relaxation. Inhibition of K+ ! 14!channels contrastingly reduces K+ conductance resulting in depolarization and constriction. Although not collected in cerebrovascular smooth muscle cells, data in mesenteric smooth muscle cells of rats in vitro indicate that a reduction in pH activates ATP sensitive potassium channels (KATP channels; Wang et al., 2003). During hypercapnia in humans, inhibition of KATP channels with glibenclamide has no effect on the blood velocity response through the MCA (Bayerle-Eder et al., 2000); however, as KATP channels directly effect vascular tone it is difficult to extrapolate velocity measures to represent volumetric CBF in this instance. Overall potassium channels regulate cerebrovascular smooth muscle tone (at least in animals) downstream of several vasoactive factors which are implicated in CO2 mediated cerebrovascular vasomotion: NO [via cyclic guanosine monophosphate (cGMP)], Adenosine, EETs, and PGs [via cyclic adenosine monophosphate (cAMP)]. Signal transduction and the role of these aforementioned vasoactive factors in CO2 mediated vascular regulation are discussed below.   1.2.3.2. Adenosine.   A potential role for adenosine in the regulation of CBF during hypercapnia has been previously demonstrated in rats as adenosine receptor antagonism via caffeine reduces the CBF response to hypercapnia (Phillis & DeLong, 1987). Moreover, the response to hypercapnia is potentiated by both the administration of dipyridamole and adenosine deaminase inhibitors, which both increase adenosine concentrations (Phillis & DeLong, 1987). In humans, treatment with caffeine (i.e., adenosine receptor antagonism) has been shown to reduce baseline CBF, but possess no effect on cerebrovascular CO2 reactivity (Blaha et al., 2007). However, as the half life of caffeine is approximately 5.5hours (Statland & Demas, 1980), and subjects were only asked to refrain from caffeine containing beverages for 6 hours, pre-existing caffeine may have contaminated baseline recordings making it difficult to draw strong conclusions from the study by Blaha et al., 2007. Therefore, the potential role of adenosine in modulating PaCO2 mediated cerebrovascular vasomotion in humans remains unclear.   ! 15!1.2.3.3. Nitric oxide.   Hypercapnia induces increased release of cGMP (Parfenova et al., 1994) a second messenger downstream of NO activity. Moreover, inhibiting nitric oxide synthase (NOS) non-selectively or by selectively inhibiting neuronal NOS (nNOS) significantly blunts the CBF response to hypercapnia in rats (Wang et al., 1994a; Smith et al., 1997), while addition of NO donors restores hypercapnia mediated increases in CBF (Iadecola & Zhang, 1996). Contrary to animal studies, some (White et al., 1998; Ide et al., 2007), but not all (Schmetterer et al., 1997) evidence in humans suggests NO may not appreciably impact cerebrovascular reactivity to hypercapnia; however, these data are based upon velocity and not flow indices of CBF. Increases in CBF in response to hypercapnia have been reported to take ~30seconds to complete and reach steady state (Shapiro et al., 1966), which may be a result of the time needed for larger cerebral vessel dilation to occur and, therefore, indicate a role for endothelial released NO. Such is the case in the periphery, where shear stress mediated dilation takes ~30seconds to occur (Black et al., 2008) - a response driven primarily by NO (Green et al., 2014). However, despite similar time dependent response characteristics between peripheral and cerebral responses it is unknown if NO is a key regulator of CBF during alterations in PaCO2.    1.2.3.4. Epoxyeicosatrienoic acids.   Metabolic byproducts of the arachidonic acid pathway called EET’s are formed via epoxygenase activity (Figure 1.4). There are four different EETs (5,6-EET; 8,9-EET; 11,12-EET; and 14,15-EET), which all vasodilate the cerebrovasculature in animals (Leffler & Fedinec, 1997), although a dilatory effect of all EETs has not been consistently demonstrated, with evidence that only 5,6-EET may be vasoactive (Ellis et al., 1990). Interestingly, the presence of prostacyclin (PGI2) seems to facilitate EET mediated dilation of pial vessels (Leffler & Fedinec, 1997). However, the direct impact of EETs on cerebral vasomotion in response to PaCO2 alterations has not been investigated in humans.   ! 16!1.3. The prostaglandin pathway In the 1930’s PGs were discovered by Kurzok and Lieb (Kurzrok & Lieb, 1930).  However, PGs were not isolated in their pure form until the mid 1960’s (Samuelsson, 2012). The presence of PGI2 was later determined in 1976 (Moncada et al., 1976). Since both peripheral (Busse et al., 1984; Nicholson et al., 2009), and cerebral (Davis et al., 2004) vascular smooth muscle relax in response to PG’s, they have been implicated as an important factor in vascular function and health. These PGs are produced as an end product of the arachidonic acid pathway (Weksler et al., 1977), which is mediated in part by cyclooxygenase (COX) activity (Smith, 1992).  This makes COX an ideal target to inhibit for the investigation of PG mediated functions, including cerebrovascular regulation. A downstream product of the arachidonic acid pathway (Weksler et al., 1977), PG production results from COX conversion of arachidonic acid through to prostaglandin H2  (PGH2; Smith et al., 1991; Smith, 1992). Specifically, this pathway begins with the conversion of arachidonic acid to prostaglandin-G2 (PGG2) by COX. This PG2 is then reduced to prostaglandin endoperoxide-H2 (PGH2; also commonly referred to as PG synthase) by peroxidase. Finally, specific isomerization or reduction of PGH2 produces one of either prostaglandin-D2, E2, F2α (PGD2, PGE2, & PGF2α, respectively), PGI2, or thromboxane-A2 (Smith, 1992). Figure 1.4 depicts a simplified outline of PG production as well as the co-existing lipoxygenase and epoxygenase pathways which are likely also implicated in the control of CBF via flux through the arachidonic acid pathway (Attwell et al., 2010). The production of PG’s is seemingly endothelial dependent in both peripheral (Weksler et al., 1977; Messina et al., 1992) and cerebrovascular (Hsu et al., 1993)  blood vessels. Post mortem studies have demonstrated the ability of cerebral vascular tissue to produce PG’s, primarily PGI2 and also PGE2 to a lower extent (Abdel-halim et al., 1980), highlighting that they are present in - and likely vasoactive regulators of - the cerebrovascular circulation.  As PGI2 has a relatively short half life (~30seconds) and the more stable PG’s are quickly metabolized, it is likely that PG’s primarily mediate effects localized to their production site, thus necessitating this production via cerebral vessels (Narumiya et al., 1999). On a whole body scale, the production of PG’s is ubiquitous. The potential role for PG’s as regulators of cerebrovascular CO2 reactivity will now be ! 17!extensively reviewed. First, animal data In Vitro and In Vivo will be summarized followed by an overview of In Vitro and In Vivo work in humans.  Figure 1.4. The arachidonic acid pathway. Arachadonic acid is produced from membrane phospholipids through the activity of phospholipase A. It is then converted into one of three compounds: PGG2, epoxyeicosatrienoic acid (EET’s), and 20- hydroxyeicosatetraenoic acid (20-HETE). Inhibition of cyclooxygenase is commonly used to inhibit the production of PG’s from arachidonic acid. PG, Prostaglandin; PGG2, Prostaglandin G2; PGH2, Prostaglandin H2; PGE2, Prostaglandin E2; PGD2, Prostaglandin D2; PGI2, Prostacyclin; PGG2, Prostaglandin G2; PGH2, Prostaglandin H2; PGF2α, Prostaglandin F2α; TXA2, Thromboxane A2   1.3.1. Prostaglandins and animal cerebral vasomotor tone: In Vitro  The putative role of PG’s in the cerebrovascular regulation of animals is based on the production of PG’s In Vitro in rats (Gecse et al., 1982), guinea pigs (Gecse et al., 1982), and pigs (Parfenova et al., 1995b). Supportive evidence is also based on the presence of PG synthases (Boullin et al., 1979).  Synthesis of PG’s likely occurs in both smooth muscle (Parfenova et al., 1995a) and endothelial cells (Hsu et al., 1993). Addition of exogenous PG’s to In Vitro tissue baths, such as PGE2, PGI2 and iloprost (a synthetic PGI2 analogue) dilate the MCA and BA of cats (Whalley et al., 1989; Parsons & Whalley, 1989) – these responses are unaffected by the application of Indomethacin (INDO; Whalley et al., 1989).  ! 18!This finding is indicative of the action of these agents downstream of COX (i.e., smooth muscle receptor binding). The VA and BA of baboons have also been demonstrated to relax upon induction of PGI2 into a tissue bath (Boullin et al., 1979). While PGE2 is largely considered to be a vasodilator, there is data to the contrary in dogs, indicating PGE2 induced contraction of MCA and BA (Toda & Miyazaki, 1978) highlighting the likelihood of between species response variation. Consistent with the majority of reports exemplifying PG mediated dilation of large cerebral arteries, it has been repeatedly observed that the smaller pial vessels of the brain dilate significantly in response to numerous PG’s (Welch et al., 1974; Ellis et al., 1979). Collectively, In Vitro animal data indicate that PG’s are likely an important factor in regulating CBF, at least in these aforementioned animals studies and related preparations.   1.3.2. Prostaglandins, animal CBF and CO2 reactivity: In Vivo   Topical application of arachidonic acid  (e.g., 200µg/mL in artificial cerebrospinal fluid) stimulates increased cerebral PG production (Kontos et al., 1985; Ellis et al., 1990; Leffler et al., 1993) and subsequent pial vessel dilation in cats (Wei et al., 1980; Kontos et al., 1984), rabbits (Ellis et al., 1990), and newborn pigs (Leffler et al., 1990, 1993). Direct application of specific prostanoids, such as PGE2 and PGI2, causes dose dependent dilation of pial vessels in newborn pigs (Leffler & Busija, 1985, 1987) and cats (Wahl et al., 1973, 1989). Iloprost has also been demonstrated to induce pial arteriolar dilation in newborn pigs, with this dilation blunted by the administration of INDO (Parfenova et al., 1995b). This marked blunting of dilation in response to PGI2 receptor agonism (via iloprost) is indicative that, in addition to COX inhibition, INDO also inhibits PGI2 receptor agonism mediated signal transduction and consequent vasodilation (Parfenova et al., 1995b). Moreover, the selectivity of INDO’s effect on PGI2 receptors was confirmed when INDO was shown to have no effect on β-adrenoreceptor mediated dilation of pial vessels (Parfenova et al., 1995b). This finding is in contrast to the In Vitro studies where INDO does not affect the vasomotor response to topically applied PG’s (Whalley et al., 1989). These discrepancies highlight the potential limitations of comparing In Vitro and In Vivo data regarding PG mediated cerebral vasomotion. ! 19! Much like topical application of arachidonic acid, hypercapnia has been reported to increase cerebral PG synthesis in newborn pigs (Leffler et al., 1993; Parfenova et al., 1994) and cultured endothelial cells (Parfenova & Leffler, 1996). Inhibition of PG synthesis with the administration of INDO results in a marked reduction of cerebrovascular CO2 reactivity in baboons (Pickard & MacKenzie, 1973) and newborn pigs (Leffler et al., 1993, 1994). However, this response seems to be selective to INDO in some species as aspirin does not inhibit hypercapnic pial vessel dilation in newborn pigs (Parfenova et al., 1995b), but not in others as cyclooxygenase-1 (COX1) inhibition also blunts the hypercapnic CBF response in rats (Niwa et al., 2001). Drug dependent differences are further apparent in that INDO selectively reduces CBF in newborn pigs, whereas Aspirin, Ibuprofen, and Naproxen do not (Chemtob et al., 1991).   1.3.3. Prostaglandins and human cerebral vasomotor tone: In Vitro  Evidence for a role of PG’s, at least when using In Vitro tissue baths, in mediating cerebral vasomotion in humans was reported in 1979 due to both its endogenous production in the cerebrovasculature and vasomotor influence (Boullin et al., 1979; Hagen et al., 1979). The two most common vasodilator PG’s - PGE2 and PGI2 - are both capable of causing vasodilation of human cerebral arteries. Specifically, PGE2 and iloprost both dilate human MCA segments pre-contracted with phenylephrine (Davis et al., 2004). In addition to the PGI2 mediated dilation of the MCA, In Vitro application of PGI2 also dilates the human BA (Boullin et al., 1979; Paul et al., 1982; Parsons & Whalley, 1989), indicating that PGI2 acts as a vasodilatory agent in both the anterior and posterior cerebral circulation. However, the vasodilatory effect of PGE2 appears exclusive to the anterior circulation as it dilates MCA segments (Davis et al., 2004), but constricts BA segments (Toda & Miyazaki, 1978; Parsons & Whalley, 1989). These regional differences are likely due to differences in dish preparation, pre-contractile agents, and concentrations of PGI2 (or related analogues) used (Uski et al., 1983); such differences could also be indicative of regionally specific cerebrovascular regulation by PG’s.  ! 20!Binding of PG receptors (i.e., IP and EP4 receptors; Figure 1.5) occurs on the vascular smooth muscle cell membrane with the resultant signal transduction independent of the endothelium (Davis et al., 2004). While exogenous PG’s clearly possess vasomotor effects on the cerebral circulation as outlined above, their relevance to endogenous vasomotor regulation is limited. However, PGI2 synthase in human BA and VA’s has been demonstrated (Boullin et al., 1979), providing evidence that endogenously produced PG’s play a role in cerebrovascular regulation.   Figure 1.5. Putative pathways for prostanoid mediated signal transduction. Binding of prostacyclin and PGE2 to IP and EP4/EP2 receptors, respectively, on the vascular smooth muscle cell wall leads to increases in intracellular cAMP via stimulation of adenylate cyclase (Narumiya et al., 1999). Increases in cAMP and subsequent increases in cAMP dependent protein kinase lead to smooth muscle relaxation (Narumiya et al., 1999; Kerrick & Hoar, 1981). This is achieved via cAMP dependent protein kinase mediated phosphorylation of myosin light chain kinase, which reduces its activity (Adelstein et al., 1978), thus reducing phosphorylation of myosin and it’s binding with actin. Indomethacin (INDO) inhibits cAMP dependent protein kinase activity (Kantor & Hampton, 1978), which would lead to increased phosphorylation of myosin light chain kinase (or reduced inhibition via cAMP dependent protein kinase); therefore, increasing phosphorylation of myosin and binding with actin. Moreover, In Vivo it has been shown that INDO also reduces prostacyclin receptor (IP) mediated increases in cAMP (Parfenova et al., 1995).   ! 21!While PGI2 and PGE2 primarily possess vasodilatory roles in the cerebral circulation, their upstream substrate, PG endoperoxides (PGG2 & PGH2) cause constriction of human BA’s (Boullin et al., 1979). It is therefore plausible that cerebrovascular constriction and reductions in CBF due to PG inhibition (via INDO) are resultant from an increased PG endoperoxide concentration (from reduced flux through arachidonic acid pathway). These latter changes occur in addition to the reduced production of vasodilatory PG’s.  1.3.4. Prostaglandins and human CBF: In Vivo  Irrespective of the measurement technique used, non-selective COX inhibition with INDO causes an approximate 20-30% reduction in resting CBF. This has been exemplified in studies utilizing TCD (Markus et al., 1994; Xie et al., 2006; Fan et al., 2010), the Kety & Schmidt technique (Eriksson et al., 1983; Wennmalm et al., 1984), and varying MRI techniques (Bruhn et al., 2001; St Lawrence et al., 2002). However, this effect is generally exclusive to INDO as resting CBF is largely unaffected by a wide array of other COX inhibitors such as acetylsalicate (Eriksson et al., 1983; Wennmalm et al., 1984; Markus et al., 1994; Bruhn et al., 2001; Table 1.0), naproxen (Eriksson et al., 1983; Wennmalm et al., 1984), and sulindac (Markus et al., 1994). A recent contrasting study reported a reduction in baseline CBF in response to Naproxen; however, this study used peak systolic MCAv as its index of flow, which is a poor indication of mean velocity over an entire cardiac cycle (Szabo et al., 2014). Infusion of epoprostenol (PGI2) counter-intuitively causes a slight reduction in CBF (~8%), however, it is unclear if this effect was due to the concurrent reductions in MAP (Brown & Pickles, 1982; Pickles et al., 1984) versus that of a direct PGI2 mediated vasoconstriction.          ! 22!Table 1.0. Cyclooxygenase (COX) inhibitors used to investigate the role of prostaglandins on cerebral blood flow.  Drug COX-1 COX-2 Indomethacin ✔ ✔ Naproxen ✔ ✔ Ketorolac ✔ ✔ Ibuprofen ✔ ✔ Acetylsalicylic Acid ✔  Sulindac ✔ ✔ Celebrex  ✔ Cyclooxygenase, an important enzyme responsible for prostanoid synthesis has two specific isoforms, COX1 and COX2. Drugs differ as to whether they inhibit only one isoform, or both, or in their relative affinity for each COX isoform (Brune & Patrignani, 2015). This table summarizes in a simplistic nature which isoform previously used drugs inhibit.  1.3.5. Prostaglandins and human cerebrovascular CO2 reactivity: In Vivo  Cerebrovascular CO2 reactivity is reduced by approximately 40-60% following INDO (Wennmalm et al., 1981; Eriksson et al., 1983; Xie et al., 2006; Hoiland et al., 2014; Figure 1.6), with impairments likely greater in grey matter as opposed to white matter (St Lawrence et al., 2002). Similar to the effect reported on resting CBF, reductions in cerebrovascular CO2 reactivity following COX inhibition are exclusive to the use of INDO, despite other drugs (i.e., aspirin & naproxen) efficaciously inhibiting PG synthesis (Eriksson et al., 1983). Upon chronic use of INDO (e.g., three days) CO2 reactivity is restored to normal (Pickles et al., 1984). However, chronic treatment with INDO for one week has been contrastingly shown to reduce CO2 reactivity (Eriksson et al., 1983). Whether this difference over time (i.e., three days versus one week) is a result of methodological differences between studies or a time dependent effect of INDO is unclear. Due to the unique ability of INDO to reduce CO2 reactivity it has been proposed - although not clearly established - that INDO induced impairments in resting CBF and CO2 reactivity are due to a mechanism independent of PG synthesis inhibition (Eriksson et al., 1983; Markus et al., 1994). Despite a differential effect on CBF when compared with other COX inhibitors (i.e., Aspirin, Ibuprofen, Naproxen, etc.; Table 1.1), INDO completely inhibits platelet aggregation in response to arachidonate ! 23!application similar to both aspirin and naproxen indicating that its level of PG synthesis inhibition does not differ from other COX inhibitors (Eriksson et al., 1983). Whilst an alternate mechanism, if indeed present, remains to be elucidated, there is evidence that INDO effects cerebrovascular CO2 reactivity by reducing brain tissue extracellular pH mediated signal transduction in rats (Wang et al., 1993) or by inhibiting smooth muscle cyclic adenosine monophosphate (cAMP) and cAMP dependent protein kinase activity (Kantor & Hampton, 1978; Parfenova et al., 1995b), or both.    Figure 1.6. Absolute cerebrovascular CO2 reactivity during hyperoxic hypercapnia pre- and post 1.45mg/kg Indomethacin in humans. Individual (○) and mean (■) values. Mean cerebrovascular data plotted against steady-state iso-oxic hypercapnia steps (0, +3, +6 and +9 PETCO2). A. Absolute MCAv responses pre- and post-INDO (cm/s ⋅ mmHg PETCO2-1; n=13); B. Absolute PCAv responses pre- and post-INDO (cm/s ⋅ mmHgPETCO2-1); C. Absolute ICA responses pre- and post-INDO (ml/min ⋅ mmHgPETCO2-1); D. Absolute VA responses pre- and post-INDO (ml/min ⋅ mmHg PETCO2-1). *denotes a significant change from baseline post INDO administration (P<0.05). Reproduced from Hoiland et al., 2015 with ©permission.   Assessing the potential role of PG’s in mediating cerebral vasodilation during alterations in PaCO2 is a complex task for two main reasons. First, PG production results from flux through ! 24!the arachidonic acid pathway, a pathway that produces three different end substrates; PG’s (via COX); 20-hydroxyeicosatetraenoic acid (20-HETE; via lipoxygenases); and EET’s (via epoxygenases). The potential for changes in expoxygenase and lipoxygenase activity during inhibition of COX, due to increased substrate availability, poses as a difficulty (especially in humans) when attempting to specifically partition the role of PG’s from other vasoactive factors. Both 20-HETE and EET’s are vasoactive (Attwell et al., 2010), possessing constrictive (Wagerle & Mishra, 1988) and dilatory (Leffler & Fedinec, 1997) effects, respectively. Second, the inhibition of COX has a tendency to be ineffective in reducing cerebrovascular CO2 reactivity unless inhibited by INDO as previously stated. Other non-selective COX inhibitors (i.e., ibuprofen, naproxen) have no effect on CBF reactivity to changes in PaCO2 (see Table 1.1). Thus, it is imperative to consider both the complexity of the signaling pathway and specific pharmacology when assessing the role of PG’s in mediating cerebrovascular CO2 reactivity and vasomotion. Additional consideration of both the arachidonic acid pathway and several COX inhibitors and their effect on cerebral vasodilation are now provided. This includes discussion of vasoactive factors such as reactive oxygen species (ROS) and NO, as well as pharmacological effects on cell signaling and Ca2+ sensitivity.    ! 25!Table 1.1 The effect of prostaglandin synthesis inhibitors on resting cerebral blood flow and cerebrovascular CO2 reactivity in humans.  Drug Study Population Measurement Technique Δ CBF - Rest Δ Reactivity- Hypercapnia Δ Reactivity - Hypocapnia Indomethacin (COX 1 & 2)       100mg (oral) Markus et al., Stroke, 1994 Both Sexes (33±6years) TCD -30%* -34% (NS) -45%* 100mg (oral) Kastrup et al., J Neurol Sci, 1998 Females (28±2years) TCD -17%* -66%*  100mg (oral) Kastrup et al., J Neurol Sci, 1998 Males (31±4years) TCD -19%* -56*  0.2mg/kg (intravenous) St. Lawrence et al., J Magn Reson Im, 2002  Both Sexes (35±8years) MRI (Arterial Spin Labeling) -36%* -63%*  0.2mg/kg (intravenous) Bruhn et al., J Magn Reson Im, 2001 Both Sexes (range: 22-31, mean: 26 years) MRI (Blood Oxygen Level Dependent) -40%   1.5mg/kg (oral) Eriksson et al., Acta Physiol Scand, 1983 Males (range: 21-32years) Kety-Schmidt Technique -20%* -54%*  0.8mg/kg (oral) Wennmalm et al., Arch Toxicol Suppl, 1984 Male (range: 20-32years) Kety-Schmidt Technique -35%e -50%e  100mg (rectal) Jensen et al., J Neurosurg Anesth, 1996 Both Sexes (range: 18-48years) 133Xe scintillation detectors -35%*   100mg (rectal) Wennmalm et al., Clin Physiology, 1981 Male (range: 20-40years) Kety-Schmidt Technique -35%* -72%*  25-50mg (oral) Okabe et al., J Cereb Blood Flow & Metab, 1983 Both sexes (44±15years) 133Xe scintillation detectors -21%*   Aspirin (COX 1)       1200mg (oral) Markus et al., Stroke, 1994 Both Sexes (35±8years) TCD -7% (NS) 5% (NS)e 5% (NS)e 0.5g over 2 minutes (intravenous) Bruhn et al., J Magn Reson Im, 2001 Both Sexes (range: 22-31, mean: 26 years) MRI (Blood Oxygen Level Dependent) No change   45mg/kg (oral) Eriksson et al., Acta Physiol Scand, 1983 Both Sexes (range: 21-31years) Kety-Schmidt Technique -7% (NS) -1% (NS)  45mg/kg (oral) Wennmalm et al., Arch Toxicol Suppl, 1984 Male (range: 20-32years) Kety-Schmidt Technique No change 9% (NS)  Naproxen (COX 1 & 2)       4mg/kg (oral) Eriksson et al., Acta Physiol Scand, 1983 Males (range: 24-48years) Kety-Schmidt Technique -1% (NS) -8% (NS)e  4mg/kg (oral) Wennmalm et al., Arch Toxicol Suppl, 1984 Male (range: 20-32years) Kety-Schmidt Technique -8% (NS) -14% (NS)  Sulindac (COX 1 & 2)       300mg (oral) Markus et al., Stroke, 1994 Both Sexes (34±8years) TCD 1% (NS) <1% (NS) -10% (NS) Data are arranged to show the effect of multiple prostaglandin synthesis inhibitors across studies using various measurement techniques. Collectively, the presented data highlight that only indomethacin is capable of significantly reducing CBF and cerebrovascular CO2 reactivity. Due to several studies failing to report specific values for drug effects they have been estimated from presented graphs and figures. These studies have been noted with an e symbol. ! 26!1.3.6. Prostaglandin pathway: cyclic adenosine monophosphate production  In vitro, both cerebral microvascular smooth muscle and endothelial cells produce low levels of cAMP (Parfenova & Leffler, 1996) with PGI2 and PGE2 application markedly increasing cAMP production/concentrations above basal levels (Parfenova et al., 1995a). Topically applied cAMP results in pial arteriolar dilation in newborn pigs, with pial arterial dilation strongly related to the increases in cAMP concentration (r>0.80; Parfenova et al., 1993). Moreover, hypercapnia (Parfenova et al., 1993, 1994), and both PGE2 (Parfenova et al., 1995a) and PGI2 receptor agonism (Parfenova et al., 1994, 1995b) increase cerebrovascular cAMP production. These findings provide compelling evidence that hypercapnia induced increases in cAMP are in part PGE2/PGI2 receptor agonism dependent (see Figure 1.5). This likely dependence of cAMP production during hypercapnia on PG activity (i.e. IP or EP receptor binding), is apparent in that INDO concurrently reduces cAMP production and vasomotor responses subsequent to PGI2 receptor agonism (via iloprost) by 50-75% (Parfenova et al., 1995b). Parfenova et al., 1995 were able to more specifically highlight that the reduced cAMP production is a direct consequence of PGI2 receptor agonism and not COX inhibition, as aspirin (also a COX inhibitor) does not inhibit PGI2 receptor mediated vasomotor responses and had no effect on iloprost induced cAMP production (Parfenova et al., 1995b). Thus, where other COX inhibitors have no effect, it seems INDO has a specific effect (that is not COX related) relative to inhibition of PGI2 receptor mediated responses, potentially explaining its potent ability to inhibit cerebrovascular responses to CO2. However, the exact location of this effect in the post-receptor signaling cascade is not clear. Thus, there is likely some effect of INDO on post receptor activation of adenylate cyclase or adenylate cyclase mediated production of cAMP, or both. These effects are independent to the downstream INDO induced inhibition of cAMP dependent protein kinase (Kantor & Hampton, 1978) which acts to increase smooth muscle cell Ca2+ sensitivity and blunt vessel dilation (Figure 1.5).    While increased production of cAMP is a consequent effect of PGI2 receptor agonism via iloprost, hypercapnia also induces increases in cerebral cAMP (Parfenova et al., 1993, 1995b). This PGI2 receptor mediated production of cAMP during hypercapnia is an attractive ! 27!hypothesis to explain hypercapnic vasodilation as reductions in pH may increase PGI2 receptor affinity (Hashimoto et al., 1990), making PGI2 a potent vasoactive agent in an acidic milieu. It is plausible that only an extremely low amount of PGI2 is necessary to trigger this hypercapnic PGI2 receptor mediated production of cAMP. For example, near complete (i.e. >70-80%) inhibition of PGI2 production via Aspirin has no effect on cAMP production subsequent to iloprost infusion or hypercapnia in newborn pigs (Parfenova et al., 1995b) and PGI2 mediated facilitation of hypercapnic dilation is not dose dependent (Leffler et al., 1999). Therefore, PG’s may still be mediating dilation despite apparent abolished synthesis with other COX inhibitors. However, whether only a low level of PGI2 is necessary to elicit hypercapnic dilation, or there are PG independent factors increasing cAMP (Parfenova & Leffler, 1996) in vivo remains unknown. Taken together, the production of cAMP during hypercapnia, and inhibition of PGI2 receptor mediated cAMP production via INDO, likely explains how INDO reduces cerebrovascular CO2 reactivity in animals (Pickard & MacKenzie, 1973; Parfenova et al., 1995b) and humans (Eriksson et al., 1983).  Therefore, despite the apparent inability to detect hypercapnia induced PG synthesis in humans at baseline (Eriksson et al., 1983) or in animals treated to block PG synthesis (Leffler et al., 1993), PG’s may still be mediating responses at an extremely low plasma concentration.   In summary, in addition to its efficacy in inhibiting PG synthesis, INDO is likely inhibiting a post-receptor signal cascade that is reliant on only minimal PGI2 activity (Parfenova et al., 1995b). Moreover, it is important to consider that downstream of changes in cAMP production, INDO also inhibits cAMP dependent protein kinase activity (Kantor & Hampton, 1978). These finding suggest that INDO may be acting in two additional manners compared to other COX inhibitors: 1) a prostaglandin dependent mechanism whereby IP receptor stimulation by PGI2 increases smooth muscle cell cAMP levels, independent of COX inhibition, and 2) a prostaglandin independent mechanism whereby the inhibition of cAMP-dependent protein kinase results in increased smooth muscle cell calcium sensitivity.     ! 28!1.3.7. Prostaglandin pathway: reactive oxygen species production  Increased PG production, induced via topical application of arachidonic acid within a cranial window, is associated with ROS production in cats (Kontos et al., 1985). The production of ROS resulting from the application of arachidonic acid is inhibited to the same extent by superoxide dismutase and INDO, suggesting that the increase in ROS is mediated via COX activity (Kontos et al., 1985). In COX1 null mice, superoxide dismutase has no effect on bradykinin induced dilation, indicating COX1 is responsible for ROS release during flux through the arachidonic acid pathway in this species (Niwa et al., 2001). In addition to their similar effects on ROS production, superoxide dismutase and INDO equally block vasodilation to both 5,6-EET and arachidonic acid (Ellis et al., 1990). This 5,6-EET is a COX substrate providing further evidence that COX produced ROS results in cerebral vasodilation in both cats and rabbits (Ellis et al., 1990). While the implications of concurrent ROS production during PG synthesis in regulating cerebrovascular tone is unknown in humans, it has been shown that ROS inhibition with superoxide dismutase and catalase nearly abolishes the vasodilatory response to topical arachidonic acid application in cats (Kontos et al., 1984) and rabbits (Ellis et al., 1990). Further, pial arterioles of cats (Wei et al., 1985), mice (Rosenblum, 1983) and newborn pigs (Leffler et al., 1990) dilate in response to increased ROS production. Taken together, at least at rest in animals, these studies indicate that pial vessel dilation in response to flux through the arachidonic acid pathway is largely due to ROS production secondary to COX activity.  Contrary to topical administration of arachidonic acid, combined superoxide dismutase and catalase do not attenuate the vasodilatory response to increased PaCO2 in cats (Kontos et al., 1984), newborn pigs (Wagerle & Mishra, 1988; Leffler et al., 1991) and mice (Rosenblum, 1983). However, supporting data in cats and rabbits specifically may not be relevant for hypothetical translation to humans as INDO administration in these animals also does not attenuate the CBF response to hypercapnia (Wei et al., 1980; Busija, 1983; Busija & Heistad, 1983) signifying hypercapnic vasodilation is not a PG dependent response in these species. Similar to topical application of arachidonic acid, hypercapnia also increases flux through the arachidonic acid pathway and consequent PG synthesis (Leffler et al., 1991); however, this ! 29!finding is not universal (Ellis et al., 1980). Thus reductions in hypercapnia mediated dilation in response to treatment with COX inhibitors are likely attributable to a reduction in PG’s but not the concurrent reduction in COX mediated ROS production (Leffler et al., 1991). For example, superoxide dismutase has no effect on the CBF response to hypercapnia in both normal and COX1 null mice (Niwa et al., 2001). Interestingly, superoxide dismutase greatly reduced (~50%) the cerebral dilation response to topical application of bradykinin (a peptide that causes downstream PG production and COX activity, in normal mice, but not COX1 null mice). These findings, in combination with the lack of effect during hypercapnia in both groups of mice, highlights that whilst ROS are an important mediator of dilation in conjunction with COX activity, this mechanism likely has no appreciable effect during hypercapnic challenges (Niwa et al., 2001). Further, it is unlikely that ROS effect hypercapnia induced PG production (Leffler et al., 1991). These studies highlight that the effects of ROS on vascular tone during increased PG synthesis at rest cannot be extrapolated to hypercapnic challenges, as COX mediated hypercapnic dilation is seemingly independent of the mediation of vasomotor tone by ROS in a multitude of species. This phenomenon remains to be studied in humans.   1.3.8. Prostaglandin pathway: interaction with nitric oxide  As both PG and NO formation are predominantly endothelium dependent they are commonly thought to interact. However, little evidence exists as to the potential synergistic or facilitative effects of NO and PG’s in the cerebrovascular response to changes in PaCO2. As hypercapnia leads to increases in cGMP a second messenger in the NO signal transduction pathway (Parfenova et al., 1993, 1994) concurrent to the aforementioned increases in cAMP this highlights the potential for these signaling pathways to modulate the other. Moreover, while iloprost and PGE2 lead to increases in cerebral endothelial production of cGMP (Parfenova et al., 1995b), INDO markedly attenuates the cGMP response to hypercapnia (Parfenova et al., 1994). This potential interaction is also apparent in that INDO inhibits the hyperemic effect of applied L-arginine in rats during hypercapnia (Wang et al., 1994b).  While the lack of data in animals precludes any conclusion as to potential NO and PG interactions during hypercapnia, the extremely limited study of NO in the regulation of ! 30!human cerebrovascular CO2 reactivity (White et al., 1998; Ide et al., 2007) highlights how little is known as to the interaction of these systems in healthy humans.   1.4. Measurement of cerebral blood flow  Although considerable anatomical knowledge was acquired by anatomists three centuries ago, little was known about the brain’s circulatory control until the latter half of the nineteenth century. At this time, the first measurements of CBF were made by the Italian physiologist, Angelo Mosso. Mosso, in a select population of patients with skull defects, was able to measure cerebral volume changes, or brain pulsations, in response to emotional and cognitive stimuli, subsequent to his development of a specialized plethysmograph (Mosso, 1880). This provided the first insight into cerebrovascular regulation in humans, and signifies the starting point in the development of a wide array of CBF measurement techniques occurring over the last ~150 years. Some 50 years after Mosso’s original work, Kety and Schmidt developed an invasive method incorporating nitrous oxide as a tracer to measure CBF (Kety & Schmidt, 1948a). This technique was predominant in the cerebrovascular research world until Aaslid and colleagues, in 1982, demonstrated the utility of the TCD for the non-invasive assessment of CBF (Aaslid et al., 1982). This technique, which has been used across multiple fields for the assessment of numerous pathologies (i.e., stenosis, anemia, brain death, cerebral vasospasm, etc.), has largely shaped contemporary knowledge of cerebrovascular physiology.  Within the past decade, more advanced ultrasound techniques have been introduced into the field of cerebrovascular physiology. These techniques include both duplex ultrasound, and transcranial colour coded Doppler.  Detailed discussions on other advanced imaging techniques such as MRI and PET are beyond the scope of this thesis and are reviewed elsewhere (Rota Kops et al., 1990; Zwanenburg et al., 2011). Since this thesis is based upon the use of TCD and peripheral duplex ultrasound, the primary principles of these techniques will now be discussed in detail.    ! 31!1.4.1. Transcranial Doppler ultrasound.   The utility of Doppler ultrasound for the measurement of blood velocity in peripheral blood vessels was reported as early as 1965 (Miyazaki & Kato, 1965), while its applicability within the cerebral circulation was first described in 1982 (Aaslid et al., 1982). Aaslid and colleagues (1982) demonstrated that thin areas of the skull (or “acoustic windows”) allow for transmission of low frequency (2MHz) ultrasound for the insonation of various intra-cranial vessels. Thus, measurement of cerebral blood velocity in the large intra-cranial cerebral arteries of the brain (ACA, MCA, PCA & BA) was made possible in real time and on a beat-to-beat basis and, therefore, at a much greater temporal resolution than previous steady state techniques allowed for (Kety & Schmidt, 1948a). The primary limitation of TCD is that it does not provide an absolute measurement of volumetric blood flow, but rather measures the speed of red blood cells travelling through the insonated vessel. Therefore, the potential for changes in arterial diameter during alterations in PaO2 (Wilson et al., 2011), PaCO2 (Giller et al., 1993; Ainslie & Hoiland, 2014; Verbree et al., 2014; Coverdale et al., 2014, 2015; Figure 1.7), and blood pressure (Lewis et al., 2015) must be accounted for when interpreting TCD data. For example, traditionally it had been assumed that the MCA does not change in diameter during alterations in PaCO2, with this assumption largely based upon a low resolution (1.5T) magnetic resonance imaging (MRI) study performed in the year 2000 by Serrador and colleagues. However, recent high resolution (3 & 7T) MRI and ultrasound data have contested these previous assumptions, making it apparent that MCA diameter does change during PaCO2 perturbations greater than approximately 7.5mmHg above or below baseline values (Willie et al., 2012; Verbree et al., 2014; Coverdale et al., 2014, 2015). Thus, the use of TCD will lead to underestimation of flow during hypercapnia, and overestimations of flow during hypocapnia. Such limitations have encouraged the measurement of CBF via Duplex ultrasound through the extra-cranial blood vessels (ICA & VA). This approach permits the quantification of both velocity and diameter of the vessel of interest.  ! 32! Figure 1.7. The impact of MCA vasomotion on the discrepancy between volumetric and velocity indices of CBF. Previously reported changes in middle cerebral artery (MCA) diameter (left y-axis) and their calculated impact on the discrepancy between flow and velocity measures (right y-axis) during changes in end-tidal PCO2 (PETCO2). To highlight the effects of MCA vasomotion we estimated the potential difference between CBF and velocity changes using the following. For example, cross-sectional area (CSA; cm2) * Velocity (cm/s) * 60s = Flow (mL/min). Assuming a baseline MCA velocity of 60cm/s (which was done for all studies to facilitate diameter effect comparisons) coupled with the observed alterations in CSA with hyper- or hypocapnia (Coverdale et al., 2014), we calculated a representative baseline MCA flow value: 5.6mm2 * 60cm/s *60 s = Flow; therefore 0.056cm2 *60cm/s * 60s = 201.6 mL/min. Assuming previously reported values of cerebrovascular reactivity (Willie et al., 2012), MCA velocity increases ~4%/mmHg increase in PETCO2. Assuming this as vessel reactivity (for all studies), we can estimate the volumetric MCA flow during hypercapnia using the reported CSA: 6.5mm2 * 84cm/s * 60s = Flow; therefore, 0.065 cm2 * 84 cm/s * 60s = 327.6 mL/min. As such, the percent difference for flow between baseline and hypercapnia is [(327.6-201.6)/201.6]*100 = 62.5%. The percent difference in velocity between baseline and hypercapnia is [(84-60)/60]*100 = 40%, indicating that TCD would underestimate the increase in flow of the MCA during hypercapnia (+9mmHgPETCO2 from baseline) by >20%. However, if the percent difference is quantified via the magnitude of change in flow and velocity during hypercapnia, the increase in flow is ~50% greater than that of velocity! This can be calculated as %difference = [(%increase in flow - %increase in velocity)/%increase in velocity] * 100 and therefore, [(62.5-40)/40]*100 = 56.25%. As such, we have conservatively represented the effect of changes in diameter on flow vs. velocity discrepancies. For hypocapnia we again assumed a baseline MCA velocity of 60cm/s and used the pre-hypocapnia CSA reported (Coverdale et al., 2014) to calculate baseline flow: 5.8mm2 * 60cm/s * 60s = Flow; therefore, 0.058cm2 * 60cm/s * 60s = 208.8 mL/min. Incorporating a 2% change in MCAv per mmHg reduction in PETCO2 (Willie et al., 2012) and the associated change in CSA, we estimated volumetric MCA flow during hypocapnia: 5.3mm2 *46.8cm/s *60s = Flow; therefore, 0.053cm2*46.7cm/s*60s = 148.8mL/min. As such, the percent change in flow between baseline and hypocapnia is [(148.8-208.8)/208.8]*100 = -28.7%. The percent difference in velocity between baseline and hypocapnia is [(46.8-60)/60]*100 = -22%, indicating that TCD would underestimate the decrease in flow of the MCA during hypocapnia (-13mmHgPETCO2 from baseline) by ~7%. Thus, it is evident by these calculations and those seen above that small changes in MCA diameter are responsible for large discrepancies between flow and velocity measures. Data are collated from (Valdueza et al., 1997; Serrador et al., 2000; Verbree et al., 2014; Coverdale et al., 2014). As noted in the hypercapnic calculations, this graph represents the most conservative way to quantify the percent difference in flow and velocity changes, highlighting the the large impact that changes in MCA have in quantifying CBF. Figure and text from (Ainslie & Hoiland, 2014). ©Permission not required. ! 33!1.4.1.1. Doppler physics.   Doppler ultrasound functions on the principle that the velocity of red blood cells is directly proportional to the magnitude of the Doppler shift. This principle means that as pulses are sent out from the ultrasound probe, the difference in time between the transmitted signal and the returning signal between two separate pulses is indicative of red blood cell velocity. For example, in the instance that blood is flowing towards the ultrasound probe, a second of two pulses will have a shorter signal transmission and signal return time than that of the original first pulse. The term “Doppler shift” signifies the difference between pulses.  Acquisition of correct velocity measures is dependent on the angle of insonation. Specifically, measures must be taken with an angle of 0-60° between the incident beam and direction of blood flow. A perpendicular beam in Doppler mode would result in Doppler shifts that are too low, poorly formed velocity waveforms, and / or incorrect determination of velocity. Therefore, it is important to approach a blood vessel so that the angle between blood flow and the incident beam is minimized. It is in this respect that proper technique for locating cerebral blood vessels is paramount in accurately measuring cerebral blood velocity with TCD.  1.4.1.2. Vessel location and insonation.   The large intra-cranial cerebral vessels can be insonated using three primary techniques: (1) the trans-temporal technique, where the probe is placed superior to the zygomatic arch; (2) the trans-occular technique, where the probe is placed over the closed eye; and (3) the foramen magnum approach, in which the probe is placed on the posterior portion of the head immediately inferior to external occipital protuberance (Willie et al., 2011a). These varying probe locations allow for proper signal optimization dependent upon the vessel of interest. To locate the MCA, the intracranial vessel pertaining to this thesis study, the trans-temporal approach is used.   ! 34!1.4.1.3. The trans-temporal approach.   Proper determinination of absolute MCAv is dependent upon correct insonation angle (Willie et al., 2011a). The transtemporal technique includes three different approaches for the insonation of the MCA: 1) through the posterior window, where the probe location is directly anterior to the zygomatic arch and immediately rostral of the pinna, 2) through the anterior window, where the probe is placed above the anterior process of the zygomatic arch, and 3) the middle window, which lies between the posterior and anterior window. Ideally the MCA is insonated through the middle window as in the absence of anatomical abnormalities this approach provides the lowest insonation angle and thus, most accurate measure of absolute blood velocity.  1.4.1.4. Quantification of cerebral blood velocity from transcranial Doppler ultrasound.   Velocity indices from TCD represent the entire range of blood velocity values throughout a vessel lumen. If blood flow is laminar the resultant blood velocity across the vessel lumen is roughly parabolic in shape, with the highest velocity occurring in the center of the vessel. A parabolic velocity profile allows for simple and accurate calculation of mean blood through a vessel as one half the peak velocity waveform envelope, or one half the time averaged peak velocity (Evans, 1985). Analysis error may be present in cases where blood velocity through the center of the vessel (as would be present with a parabolic velocity profile) is not the maximum velocity in the vessel, although it is rare that maximum velocity does not occur in the center of the vessel (Evans, 1985).  1.4.2. Duplex ultrasound of extra-cranial cerebral blood Vessels  Peripheral duplex ultrasound allows for the assessment of cerebral blood velocity using the same principles described above (see “1.4.1.1. Doppler Physics”) whilst simultaneously assessing arterial diameter. Simultaneous measurement of arterial diameter and blood velocity, in contrast to TCD, allows for the calculation of volumetric CBF. The primary advantage of this technique is that interpretation of duplex ultrasound data is not dependent ! 35!upon the assumption that arterial diameter is constant through perturbations in PaCO2.  Peripheral duplex ultrasound can be effectively utilized to measure CBF in the extra-cranial cerebral vessels of the neck, the ICA and VA (Willie et al., 2012).   In previous studies (Willie et al., 2012; Lewis et al., 2014a, 2014b),  global CBF has been calculated as double the sum of unilateral VA and ICA flow measures. When calculating CBF from unilateral ICA and / or VA measures it is important to be aware of bilateral differences as well as sex differences. For example, Seidel et al., 1999 performed a study on patients hospitalized in the neurological department and reported a significantly higher flow volume in the VA of males compared with females (Seidel et al., 1999). However, their sample size was relatively small to establish normative values (n = 50) and the authors also reported an angle of insonation of 62 ± 6° (range 45° to 75°) which is suboptimal for accurate velocity measurement. In contrast, Schoning et al., 1994 reported no difference in VA flow between men and women in healthy volunteers, indicating that differences are not present in healthy populations (Schöning et al., 1994). This has recently been corroborated by MRI measures (Zarrinkoob et al., 2015). It has also been reported that the left VA has higher flow than the right VA (Schöning et al., 1994; Seidel et al., 1999). These studies did not report any bilateral differences in flow through the ICA (Zarrinkoob et al., 2015).  1.4.2.1. Location and insonation of the internal carotid artery.   When aiming to locate the ICA, it is typical to first obtain a cross-sectional view of the common carotid artery. While maintaining both the same probe position and the cross sectional image in the center of the B-mode image, a 90° clockwise turn of the transducer will bring the common carotid artery into a longitudinal view (Figure 1.8-A). From here, by tracking the common carotid artery caudally the common carotid bifurcation will come into view. At this stage of the location process, several steps are necessary to differentiate between the external carotid artery and the ICA:  First, when viewing the common carotid bifurcation in the frontal plane it is typically observed that the ICA branches upwards and then curves downwards, while the external ! 36!carotid artery branches off downwards. In some cases, both vessels can be seen from this point of view; if not, an upwards or downwards “sweeping” motion with the transducer will alternate visualization of each vessel. Second, at the common carotid bifurcation the diameter of the external carotid artery is typically smaller than that of the ICA. As such, the ICA tends to visually appear larger. Third, the ICA has no extracranial branches. Therefore, the ability to visualize branching off from a vessel distal of the carotid bifurcation is indicative that the insonated vessel is the external carotid artery. Fourth, the velocity waveform of the ECA and ICA are distinctly different. The external carotid artery typically has a narrower systolic peak than the ICA and overall greater pulsatility index.Collectively, combining these steps will ensure that the correct vessel (in this case the ICA) is being insonated.   1.4.2.2. Quantification of CBF from duplex ultrasound.   By simultaneously collecting diameter and velocity measures via B-mode and Doppler signal acquisition, respectively, one can calculate volumetric flow by multiplying the velocity of blood flowing through the insonated vessel and the cross sectional area of the vessel. Vessel cross sectional area is calculated using the equation:  Cross sectional area = πr2  Whereby ‘r’ represents the vessel radius. In the case of duplex ultrasound, the following formula is used:  Cross sectional area = π (0.5 ⋅ d)2  Where ‘d’ represents the vessel diameter as measured by duplex ultrasound B-mode imaging. As in a vessel which flow is laminar, blood velocity is parabolic in nature across the vessel lumen. Thus, mean velocity can be simply and accurately calculated as one half of the peak velocity blood velocity through the vessel (see “1.4.1.4. Quantification of Cerebral Blood Velocity from Transcranial Doppler Ultrasound” for a more in depth explanation). Thus, volumetric blood flow can be calculated by the following equation (Evans, 1985): ! 37! Volumetric blood flow = (π (0.5 ⋅ d)2) ⋅ ((1/2)(vmax))  Where ‘vmax’ represents the maximum blood velocity through the vessel. Thus, by simultaneously measuring vessel diameter, and blood velocity, calculation of beat-by-beat blood flow is possible with adequate analysis software (see “1.3.2.3. Analysis Software”). Other parameters pertinent to vascular function can also be determined with duplex ultrasound such as shear rate, flow mediated dilation, and vessel compliance. In summary, it is important to optimize both the B-mode image of the vessel and blood velocity waveform for the assessment of beat-by-beat blood flow. Insonation angle of the vessel is important for determining a true velocity (i.e., avoiding underestimation), while changes in steering (angle) of the ultrasound beam are used to assess blood velocity parallel to the vessel walls, and therefore, in line with the direction of blood flow using an automated correction angle of 60°. Further, angling of the transducer (i.e., heel / toe maneuver) provides another technique to ensure acquisition of a proper velocity measures. Optimization of the velocity signal as outlined here typically follows optimization of the B-mode image.  ! 38! Figure 1.8. Example duplex ultrasound image of the ICA. A. A standard image of the ICA as noted in the red box, which corresponds to the anatomical area of insonation depicted on the human figure. The concurrent velocity measure is directly below the B-mode image. B. The same ultrasound image as panel A, as it would appear in the blood flow analysis software. The yellow rectangles represent example regions of interest for sampling ICA diameter and velocity, with the dotted lines representing example edge detection tracking by the software.   1.4.2.3. Analysis software.   To analyze both diameter and velocity changes during CO2 perturbations, while reducing/eliminating observer bias and/or error, it is necessary to use automated edge detection software. The analysis software “FMD/BloodFlow Software Version 4.0” (LabView 10.0) has been previously shown as highly effective in both validly (as assessed through measurement of phantom models) and reliably determining arterial diameter (Woodman et al., 2001) and blood flow responses (Black et al., 2008). Using this specific analysis software is simple and requires few steps making it a viable tool for the quantification of CBF.  ! 39!During the experimental procedure, ultrasound images must be screen captured and saved for offline analysis. Saved files can then later be loaded into the FMD/BloodFlow Software Version 4.0 program for analysis. Regions of interest (ROI) must be made to calibrate for both vessel diameter and blood velocity. Placing the ROI over a segment of the ultrasound image with a known length (i.e., B-mode depth scale or pulsewave mode velocity scale) allow for the calibration of both diameter and velocity values. Subsequently, an ROI is placed over the entire waveform. It is imperative to watch the ultrasound video through once prior to selecting the diameter ROI as it is important to select the most stable section. These procedures are adapted from Woodman et al., 2001.  The FMD/BloodFlow Software Version 4.0 program edge detection varies for the assessment of velocity and diameter. First, the peak envelope velocity waveform is it automatically detected due to the contrast in the velocity ROI. From the top down, the software detects the first pixel of contrast and uses this as a velocity value. For diameter, two signals are sent out from the middle of the ROI, one signal up and one down, detecting the first pixel of contrast. The resulting distance between the two points is calculated as a diameter value. Of note, many points are calculated at the same time with their average used as the diameter value at a rate of 30Hz (Figure 1.8-B). A representation of the software data output can be seen in Figure 1.9.  ! 40! Figure 1.9. Example analysis software data output. A typical output from the data analysis software is represented in this figure. As shown, values for blood flow, velocity, and vessel diameter are observed on a beat-by-beat basis.   1.4.3. Variability in assessing cerebrovascular CO2 reactivity  Measurement of cerebrovascular CO2 reactivity is subject to observer error stemming from both measurement error and technical limitations, as well as physiological variability either between trials or between days. Combined, these two factors can potentially make assessment of small differences difficult necessitating a complete understanding of potential sources of error when interpreting data. For example, as blood pressure changes influence the magnitude of cerebrovascular reactivity during CO2 perturbations (Willie et al., 2012; Regan et al., 2014), variations in the blood pressure response to CO2 may contribute to between trial or between day differences in cerebrovascular reactivity. Moreover, reactivity will change throughout the day in relation to changes in endothelial function, whereby reactivity is higher in the evening than in the morning (Ainslie et al., 2007). Through understanding of the potential factors implicated in measurement variability, it is possible to design studies with ! 41!adequate controls in place to mitigate the risk of artificial variability unrelated to study interventions (i.e., testing all subjects at the same time of day).   1.4.3.1. Measurement variability.   Pertaining to this thesis, within day and between day variation are of utmost interest. Typically for biological comparisons a coefficient of variation (CoV) of <10% signifies good reproducibility of a measure, while a CoV of 10-20% represents moderate reproducibility and is still considered acceptable (Quan & Shih, 1996). Calculation of CoV is a viable method to assess reliability in studies with as small of a sample size as five, indicating feasible use in human physiology experiments (Tian, 2006). Thus, ultrasound measurements for the assessment of reliability were collected as part of this thesis, and the reliability of assessing ICA blood flow (QICA) and MCAv will now be summarized.  Within and between day variability of resting QICA measures are illustrated in the Bland Altman plots of figure 1.10. For within day measures of QICA the test-retest (measures separated by ~3.5hours) CoV was 8.5%, while as expected the between day CoV was slightly higher at 12.5%. Importantly, there is no appreciable mean shift in resting QICA (<3mL/min) for both within and between day test-retest differences.    Figure 1.10. Bland-Altman plots of duplex ultrasound measurements of volumetric flow through the ICA. A. Within day reliability of measuring QICA at rest. B. Between day reliability of measuring QICA at rest. The red line represent the mean difference between tests for both figure A & B, while the dotted blue lines represent the upper and lower limits of agreement.   ! 42!Figure 1.11 shows Bland Altman plots of within day and between day test-retest measures of resting MCAv. The within day test-retest CoV was 7.6% with the between day CoV at 9.8%. Again there is no appreciable shift in the mean differences, which are both <3.5cm/s from zero. It is important to note that some of the observed variation will be due to differences in resting PETCO2 as normal resting PETCO2 can change within and between days. Collectively, these data are similar to that previously reported using duplex ultrasound (Lewis et al., 2015), and fall within the acceptable range for biological variability (i.e., <20%; Quan & Shih, 1996).    Figure 1.11. Bland-Altman plots of transcranial Doppler ultrasound measurements of blood velocity through the MCA. A. Within day reliability of measuring MCAv at rest. B. Between day reliability of measuring MCAv at rest. The red line represent the mean difference between tests for both figure A & B, while the dotted blue lines represent the upper and lower limits of agreement.   1.5. Experimental models to assess cerebrovascular CO2 reactivity  Three primary methods are used to assess cerebrovascular CO2 reactivity: Changes in fractional inspired CO2 (FiCO2); end-tidal forcing; and rebreathing methods. Each of these methods differs in their fundamental approach to alter PaCO2 and thus differ in their ability to assess a specific research question (Ainslie & Duffin, 2009; Fierstra et al., 2013). Moreover, the fashion in which these three methods are used varies between laboratories with specific variations being developed to further the utility of these tests. The specifics of each method pertinent to assessing cerebrovascular CO2 reactivity will be briefly reviewed.    ! 43!1.5.1. Changes in fractional inspired CO2  Elevations in fractional inspired CO2  (FiCO2) are commonly used to induce hypercapnia. This technique is technically simple in nature and only requires the use of a Douglas bag containing above ambient concentrations of CO2 (i.e., 6%) in air. However, there are several limitations associated with this technique. Variation in the ventilatory response to CO2 will directly effect the resultant magnitude of hypercapnia. For example, a higher ventilatory response will effectively lower PaCO2, while a low ventilatory response will result in a higher PaCO2; effectively the cerebrovascular response, which is driven by PaCO2, using steady state changes in FiCO2 is ventilatory dependent (Ainslie & Duffin, 2009). Thus it is difficult to precisely control the stimulus magnitude when assessing cerebrovascular CO2 reactivity by this method due to highly variable between subject ventilatory sensitivity (Hirshman et al., 1975). Implicated with this methodological limitation is the fact that the degree of ventilation will effect the end-tidal to arterial CO2 gradient, indicating that the accuracy of PETCO2 in reflecting PaCO2 may also be different between subjects. Specifically, during elevations in FiCO2, PETCO2 over estimates PaCO2 (Peebles et al., 2007), leading to an underestimation of reactivity.  1.5.2. The rebreathe method  Rebreathing is a simple, but specialized technique used to “decompartmentalize” CO2 in the body, or in other words eliminate the gradient between alveolar, arterial, brain tissue, and venous CO2. This is beneficial when assessing cerebrovascular CO2 reactivity as it eliminates the gradient between PETCO2 and PaCO2 and thus the cerebrovascular CO2 stimulus is known. However, the rebreathing technique precludes achievement of steady state measurements. As the full extent of CBF increases in response to CO2 are not immediate (Shapiro et al., 1966), it has been commonly demonstrated that rebreathing reactivity underestimates that of both end-tidal forcing (Pandit et al., 2003, 2007) and FiCO2 changes (Fan et al., 2010), with little data to the contrary (Brothers et al., 2014). Therefore, despite the utility of this technique for assessing other physiological parameters (i.e., ventilatory sensitivity), it is not an ideal approach to assess the full cerebrovascular response to CO2. ! 44!1.5.3. End-tidal forcing  End-tidal forcing is a technique whereby end-tidal gases (PETO2 & PETCO2) are controlled independently of ventilation. End-tidal forcing functions by prospectively calculating the required volume of O2, CO2, and N2 in the inspirate to achieve desired end-tidal values. This is achieved through breath-by-breath determination of inspiratory and expiratory tidal volume, PETO2 and PETCO2, and the subsequent use of these values to determine the required inspirate via an error reduction algorithm (Tymko et al., 2015). As this test is ventilatory independent it allows for standardization of stimulus magnitude across subjects irrespective of their inherent variability in ventilatory sensitivity. These systems also minimize the gradient between PETCO2 and PaCO2, which greatly reduces the associated risk for underestimation of cerebrovascular CO2 reactivity that is characteristic of implementing changes in FiCO2 (see “1.5.1. Changes in fractional inspired CO2”). Due to both the relatively minimal PETCO2 to PaCO2 gradient, and ventilatory independent nature of end-tidal forcing it is an ideal scientific model to manipulate CO2 for the assessment of cerebrovascular reactivity.  1.6. Purpose and hypotheses.   As the regulation of CBF in response to changes in PaCO2 remains incompletely understood, the purpose of this thesis study was to further investigate the role of PG’s in mediating cerebrovascular responses to hypercapnia and hypocapnia. More specifically, to determine the role of PG’s in mediating vasomotion of the large extra-cranial cerebral vessels through ultrasound examination of the ICA during PaCO2 perturbations.  It was hypothesized that:  1) The ICA would dilate and constrict in response to hyper and hypocapnia, respectively;   2) Orally administering INDO would markedly reduce the cerebral vasomotor (i.e., dilation / constriction) response to both hyper and hypocapnia in the ICA.  ! 45! It was further reasoned, based upon earlier reports (Eriksson et al., 1983; Markus et al., 1994), that another non-selective COX inhibitor (Ketorolac) would have no effect on the vasomotor response of the ICA to changes in CO2. !!! !! 46!Chapter!2:!Carbon dioxide mediated vasomotion of extra-cranial cerebral arteries: a role for prostaglandins?!  1Hoiland R., 1Tymko MM., 1Bain AR., 1Wildfong K.,  1Monteleone B., and 1Ainslie PN.  1Centre for Heart, Lung and Vascular Health, University of British Columbia, Okanagan Campus, School of Health and Exercise Sciences, Kelowna, BC, Canada  2.1. Overview.  Cerebrovascular regulation during perturbations in arterial CO2 is thought to occur solely at the level of the pial vessels. However, recent evidence implicates large extra-cranial cerebral blood vessels in this regulatory process. Although the mechanisms governing CO2 mediated vasomotion remain unclear, animal and human studies support a large role of prostaglandins. Thus, we examined two hypotheses: 1) vasomotion of the internal carotid artery (ICA) would occur in response to both hyper and hypocapnia; and 2) pharmacological inhibition of prostaglandin synthesis with Indomethacin (INDO; a non-selective cyclooxygenase inhibitor) would reduce the vasomotor response of the ICA to changes in end-tidal PCO2 (PETCO2). Using a randomized single-blind placebo controlled study, subjects (n=10) were tested on two occasions. Before and 90-minutes following either oral INDO (1.2mg/kg) or placebo capsule, concurrent measures of beat-by-beat blood flow, velocity and diameter of the ICA were made at rest and during steady state stages (4 min) of iso-oxic hypercapnia (+3, +6, +9mmHg above baseline) and hypocapnia (-3, -6, -9mmHg below baseline). End-tidal forcing was employed for the control of blood gases. To examine if INDO affected ICA vasomotion in a cyclooxygenase inhibition independent manner, a subset of subjects (n=5) were tested before and 45-minutes following oral Ketorolac (0.25mg/kg). During pre-drug testing in the INDO trial, the ICA dilated during hypercapnia at +6mmHg (4.72±0.45 vs. 4.95±0.51mm; P<0.001) and +9mmHg (4.72±0.45 vs. 5.12±0.47mm; P<0.001), and constricted during hypocapnia at -6mmHg (4.95±0.33 vs. 4.88±0.27mm; P<0.05) and -9mmHg (4.95±0.33 vs. 4.82±0.27mm; P<0.001). Following INDO, dilation of the ICA was still observed at +6mmHg (4.50±0.54 vs. 4.57±0.52mm; P<0.05) and +9mmHg (4.50±0.54 vs. 4.61±0.50mm; P<0.01); however, INDO reduced the vasomotor responsiveness by ! 47!67±28% (0.045±0.015 vs. 0.015±0.012mm ⋅ mmHgPETCO2-1). In the Ketorolac condition, there was no effect of the drug on the vasomotor response to hyper or hypocapnia. We conclude that: 1) changes in PETCO2 mediate vasomotion of the ICA, 2) inhibition of non-selective prostaglandin synthesis via INDO markedly reduces the vasomotor response to changes in PETCO2; and 3) INDO may be acting via a mechanism(s) independent of cyclooxygenase inhibition to reduce CO2 mediated vasomotion.  2.2. Background.   The cerebral vasculature is highly sensitive to alterations in PaCO2.  Elevations in PaCO2 (hypercapnia) cause a reduction in cerebrovascular resistance and consequent increases in CBF, while reductions in PaCO2 (hypocapnia) cause an increase in cerebrovascular resistance and decrease in CBF (Kety & Schmidt, 1948) – this response is termed  cerebrovascular CO2 reactivity. Cerebrovascular CO2 reactivity acts to attenuate fluctuations in central pH and maintain homeostatic function (Ainslie & Duffin, 2009).  Until recently, dogma posited that hypercapnia exclusively induces small pial vessel dilation, with little to no vasomotion occurring in the larger cerebral arteries (Wolff & Lennox, 1930; Serrador et al., 2000).  This assumption persisted despite human (Giller et al., 1993) and animal studies (Heistad et al., 1978) providing evidence to the contrary. Recently, however, new insight from high resolution magnetic resonance imaging has revealed that vasomotion of the MCA occurs in response to both hyper- and hypo-capnia (Verbree et al., 2014; Coverdale et al., 2014).  Furthermore, using vascular ultrasound combined with automated edge detection analysis software, dilation of large extra-cranial cerebral arteries (e.g., VA & ICA) has also been reported (Willie et al., 2012); however, these latter findings have not been consistently demonstrated when caliper based manual diameter analysis methods are used (Sato et al., 2012; Coverdale et al., 2015).   The potential mechanism(s) whereby changes in PaCO2 result in vasomotion of larger cerebral arteries include adenosine (Phillis & DeLong, 1987), NO (Parfenova et al., 1994; Smith et al., 1997), and PG’s (Wennmalm et al., 1981; Bruhn et al., 2001; St Lawrence et al., 2002).  The latter mechanism is clearly highlighted in that administration of INDO - a ! 48!non-selective COX inhibitor - reduces basal CBF by ~20-30% and cerebrovascular CO2 reactivity by  ~50-60% (Eriksson et al., 1983; Xie et al., 2005; Fan et al., 2010; Hoiland et al., 2015). While animal studies report prostaglandin mediated hypercapnic vasodilation at the level of the small pial vessels and arterioles (Pickard et al., 1980; Busija & Heistad, 1983; Leffler et al., 1991), some data collected in post-mortem humans indicate that prostaglandin-mediated vasomotion may also take place within the larger cerebral arteries (i.e., MCA; Davis et al., 2004).  As cerebrovascular CO2 reactivity is important in stabilizing central pH, and predictive of health outcome (Portegies et al., 2014), it is imperative to understand the underlying mechanisms that regulate this response.  Although COX inhibition reduces CBF (Xie et al., 2006; Fan et al., 2010; Hoiland et al., 2015) and blunts cerebrovascular CO2 reactivity (Xie et al., 2006; Fan et al., 2010; Hoiland et al., 2015), it is unknown if this is due in part to a reduction in vasomotion of the larger extra-cranial cerebral arteries. Therefore, using a single blinded placebo-controlled and randomized design, the purpose of this study was two-fold: 1) resolve the conflicting data surrounding hypercapnic dilation and hypocapnic vasoconstriction of the ICA, and 2) determine if PG’s are implicated in CO2 mediated large cerebral artery vasomotion. We hypothesized that the ICA would dilate in response to hypercapnia, and constrict in response to hypocapnia. Moreover, it was hypothesized that COX inhibition with INDO would markedly attenuate the dilatory response to hypercapnia, and the constrictive response to hypocapnia in the ICA.  Since INDO has demonstrated a unique ability to inhibit CO2 mediated cerebrovascular responses when compared to other COX inhibitors (i.e., Aspirin and Naproxen; Eriksson et al., 1983; Markus et al., 1994), in follow up experiments, we sought to assess the effects of Ketorolac - also a potent non-selective COX inhibitor - on CO2 mediated vasomotion to determine if there is a unique influence of INDO under the conditions of our experimental approach. In this respect, we reasoned that Ketorolac would not affect the vasomotor response of the ICA to hypercapnia and hypocapnia.      ! 49!2.3. Methods.   2.3.1. Subjects.  In total, fifteen healthy young volunteers were recruited to participate in this study.  The main study, (INDO investigations) included 10 subjects (1 female) with a mean age of 23±7years, and body mass index of 22±2 Kg/m2. In the follow-up study (Ketorolac investigations), a subgroup of five subjects (all male) were examined (30±7 years; body mass index of 24±2 Kg/m2). All subjects first completed written informed consent followed by a familiarization session.  During familiarization, subjects were screened to ensure reliable neck artery (VA & ICA) ultrasound images could be attained along with intracranial (MCA & PCA) signals.  Subjects were familiarized with the remaining experimental equipment and procedures during this session.  All subjects were free of any past or present cardiorespiratory and cerebrovascular disease and were not taking any prescription drugs (other than oral contraceptives; n=1) at their time of participation, as determined by a screening questionnaire. This study was approved by the University of British Columbia Clinical Research Ethics Board and conformed to the Declaration of Helsinki.  2.3.2. Experimental protocol  On the day of experimental sessions subjects arrived to the laboratory at the same time of day having refrained from alcohol, exercise and caffeine for the previous 24 hours. Subjects were instructed to lie supine for at least 15 minutes prior to beginning the study protocol and were instrumented with the experimental equipment.   2.3.2.1. Study 1.   To investigate the role of non-selective COX inhibition via oral INDO we used a single blinded, randomized, and counter balanced placebo controlled trial requiring two laboratory visits. On each day, following baseline measurements while breathing room air, the subjects performed two CO2 reactivity tests (a hypercapnic test followed by a hypocapnic test). ! 50!Thereafter, they were orally administered 1.2mg/Kg of INDO or placebo  (sugar pill matched for weight and capsule size), and repeated the baseline measures and the CO2 tests 90-minutes later (Xie et al., 2006). This dose of INDO used has been previously shown to effectively inhibit COX activity (Eriksson et al., 1983). Test days were separated by 10±9 days. The CO2 reactivity tests are as follows:  Test 1: Progressive steady state iso-oxic elevations of PETCO2. End-tidal forcing was utilized to maintain PETCO2 and PETO2 at baseline (resting) values on an individual basis for four minutes. Upon completion of this baseline stage, PETO2 remained unchanged while PETCO2 was sequentially elevated to +3, +6, and +9mmHg above baseline, with each stage lasting four minutes. Upon completion of the +9mmHg PETCO2 stage the subject breathed room air. A end-tidal forcing system was used to manipulate PETCO2 and maintain PETO2 during the hypercapnic CO2 reactivity test as previously described (Tymko et al., 2015).  Test 2: Progressive steady state iso-oxic reductions in PETCO2. End-tidal forcing was utilized to maintain PETCO2 and PETO2 at baseline (resting) values on an individual basis for four minutes. Upon completion of this baseline stage, subjects were instructed to hyperventilate to sequentially lower their PETCO2 to -3, -6, -9mmHg below baseline values while PETO2 remained unchanged. Once adequate hyperventilation was achieved, subjects were instructed to maintain constant breathing for the remainder of the test (i.e., for -3, -6, & -9mmHg stages). To achieve precise reductions of PETCO2, the CO2 concentration of the inspirate was altered on a breath-by-breath basis to compensate for variability in ventilation. A end-tidal forcing system was used to control PETCO2 and maintain PETO2 during the hypocapnic CO2 reactivity test as previously described (Tymko et al., 2015).  2.3.2.2. Study 2.   To investigate the role of non-selective COX inhibition, using orally administered Ketorolac (Toradol), subjects attended the lab on one occasion. Following baseline measurements while breathing room air, the subjects performed the identical two CO2 reactivity tests as in Study 1 (a hypercapnic test followed by a hypocapnic test). These tests were repeated 45 min later ! 51!following orally administered Ketorolac (0.25mg/kg). Previous studies have confirmed the effectiveness on COX inhibition of this dose of Ketorolac at 45 min (peak plasma concentration) with an associated half-life of ~5-6 hours (Jung et al., 1989; Jallad et al., 1990).   2.3.3. Experimental measures.  2.3.3.1. Cardiorespiratory measures.   All cardiorespiratory variables were sampled continuously throughout the protocol at 1000Hz via an analogue-to-digital data acquisition system (Powerlab, 16/30; ADInstruments, Colorado Springs, CO).  Heart rate (HR) was measured by a 3-lead electrocardiogram (ECG), and beat-to-beat blood pressure by finger photoplethysmography (Finometer PRO, Finapres Medical Systems, Amsterdam, Netherlands).  Both PETCO2 and PETO2 were sampled at the mouth and recorded by a calibrated gas analyzer (model ML206, ADInstruments) while respiratory flow was measured by a pneumotachograph (model HR 800L, HansRudolph, Shawnee, KS). Subjects’ MAP was calculated as the mean of the reconstructed brachial waveform from the Finometer.  All data was interfaced with LabChart (Version 7), and analyzed offline. Average values for the last minute of each stage were recorded.   2.3.3.2. Cerebrovascular measures.   Blood velocity through the right MCA was measured using a 2MHz TCD (Spencer Technologies, Seattle, WA). The TCD probe was attached to a specialized headband fixation device (model M600 bilateral head frame, Spencer Technologies), and then secured into place.  The MCA was insonated through the middle trans-temporal window, using previously described location and standardization techniques (Willie et al., 2011a; see “1.4.1. Transcranial Doppler Ultrasound”).  ! 52!Blood velocity and vessel diameter of the ICA was measured using a 10MHz multi-frequency linear array vascular ultrasound (Terason T3200, Teratech, Burlington, MA).  Specifically, B-mode imaging was used to measure arterial diameter, while pulse-wave mode was used to simultaneously measure peak blood velocity. Measures of QICA were made ipsilateral to the MCA. The ICA diameter and velocity were measured at least 1.5 cm distal to the common carotid bifurcation to eliminate recordings of turbulent and retrograde flow. Great care was made to ensure that the insonation angle (60°) was unchanged throughout each test. Further, for all experimental sessions, upon acquisition of the first ultrasound image there was no alteration of B-mode gain to avoid any artificial changes in arterial wall brightness / thickness. All of the ICA recordings were screen captured and stored as AVI files for offline analysis.  This analysis involved concurrent determination of arterial diameter and peak blood velocity at 30Hz, using customized edge detection and wall tracking software designed to eliminate observer bias (Woodman et al., 2001).  No less than 12 consecutive cardiac cycles were used to determine QICA. Volumetric blood flow was subsequently calculated using the following formula:  Q!"# = Peak!Envelope!Velocity2 !x![π(!!)Diameter]!  Volumetric blood flow (ICA) and velocity (MCA) values were calculated within the final minute of each four-minute steady state stage. Cerebrovascular responses were calculated separately for the hypercapnic and hypocapnic reactivity tests. All response slopes (i.e., mm ⋅ mmHgPETCO2-1) were calculated using linear regression.  2.3.4. Statistical analysis.  All resting data were compared between conditions using a one-way repeated measures ANOVA and Tukey post-hoc tests where applicable. Changes in diameter during hyper and hypocapnia were analyzed by a one-way repeated measures ANOVA within each experimental trial with Tukey post-hoc tests. Between-condition differences in diameter for a ! 53!specific PETCO2 manipulation (i.e., pre vs. post-INDO at +9mmHgPETCO2) were analyzed using a two-tailed paired t-test. Between-condition reactivities (i.e., flow and vasomotor reactivities) were compared using a one-way repeated measures ANOVA with Tukey post-hoc tests. Comparisons between pre- and post-Ketorolac variables were made with two-tailed paired t-tests. All data are expressed as means ± SD with a priori statistical significance set at P < 0.05.   2.4. Results.  2.4.1. Resting cerebral blood flow and cardiorespiratory variables.   Resting QICA was reduced by 40±12% following INDO (257.3±57.2 vs. 151.9±36.6 mL ⋅ min-1; P<0.001), while placebo treatment had no effect (257.8±60.2 vs. 252.3±62.4 mL ⋅ min-1; P=0.55). Similarly, INDO reduced resting MCAv by 36±11% (65.5±8.6 vs. 42.0±8.8 cm ⋅ s-1; P<0.001), while placebo had no effect (62.4±12.4 vs. 60.7±14.3 cm ⋅ s-1; P=0.43). Pre-INDO resting QICA and MCAv were not different from the pre-placebo and placebo trials. Resting ventilation was unaffected by INDO; however, compared to the placebo, INDO caused a modest increase in MAP (77.0±5.6 vs. 83.8±8.7 mmHg; P<0.05) and decreased HR (57±11 vs. 50±9 bpm; P<0.01).   2.4.2. Cerebrovascular CO2 reactivity (Table 2.0).   Absolute hypercapnic QICA and MCAv reactivity were reduced by 69±20% and 59±28% following INDO, respectively, but were unaltered by placebo treatment (Figure 2.0). For both QICA and MCAv, the pre-INDO, pre-placebo, and placebo trials did not differ in their respective absolute hypercapnic reactivities. Following INDO, relative hypercapnic QICA and MCAv reactivity were reduced by 50±33% and 38±36%, respectively, but unaltered by placebo treatment. For both QICA and MCAv, the pre-INDO, pre-placebo, and placebo trials did not differ in their respective relative hypercapnic reactivities.  ! 54!Table 2.0. Cerebrovascular and blood pressure responses to CO2 before and following INDO or placebo.  Hypercapnia   Absolute Reactivity  Relative Reactivity  Units Pre-INDO INDO Pre-Placebo Placebo Units Pre-INDO INDO Pre-Placebo Placebo QICA (mL · min-1 · mmHgPETCO2-1) 18.9±7.3‡ 5.3±3.0**† 17.1±5.3‡ 15.9±3.3‡ (% · mmHgPETCO2-1) 7.3±2.0‡ 3.5±2.2**† 6.7±1.9‡ 6.2±1.2‡ MCAv (cm · s-1 · mmHgPETCO2-1) 3.2±0.7‡ 1.3±1.0**† 2.8±0.7‡ 2.9±0.7‡ (% · mmHgPETCO2-1) 4.8±0.9‡ 3.0±1.9*‡ 4.5±1.3‡ 5.0±1.8‡ VE (L · min-1) 2.7±0.8 3.1±1.0 3.0±1.2 2.9±1.3 (% · min-1) 19.8±6.1 19.9±8.1 22.1±10.6 19.7±9.7 MAP (mmHg · mmHgPETCO2-1) 0.9±0.6 0.8±0.5 0.9±0.4 1.2±0.6 (% · mmHgPETCO2-1) 1.2±0.8 1.0±0.7 1.0±0.4 1.5±0.4  Hypocapnia   Absolute Reactivity  Relative Reactivity  Units Pre-INDO INDO Pre-Placebo Placebo Units Pre-INDO INDO Pre-Placebo Placebo QICA (mL · min-1 · mmHgPETCO2) 8.2±4.2 2.2±2.0** 7.9±2.2 6.8±2.2 (% · mmHgPETCO2-1) 3.0±1.0 1.2±0.9** 3.0±0.6 2.6±0.4 MCAv (cm · s-1 · mmHgPETCO2-1) 1.6±0.4 0.4±0.5** 1.6±0.4 1.5±0.5 (% · mmHgPETCO2-1) 2.4±0.5 0.7±0.8** 2.5±0.4 2.4±0.5 MAP (mmHg · mmHgPETCO2-1) -0.4±0.5 -0.08±0.6 -0.3±0.6 -0.04±0.8 (% · mmHgPETCO2-1) -0.5±0.7 -0.07±0.7 -0.4±0.8 -0.002±0.9 * significantly lower than pre-INDO, P<0.01; ** significantly lower than Pre-INDO, P<0.001; † significant difference between hypercapnic and hypocapnic reactivities within an experimental trial, P<0.01; ‡significant difference between hypercapnic and hypocapnic reactivities within an experimental trial, P<0.001  ! 55!Although unchanged in the placebo trial, absolute hypocapnic QICA and MCAv reactivity were significantly reduced by 72±25% and 79±25%, respectively following INDO (Figure 2.1). For both QICA and MCAv, the pre-INDO, pre-placebo, and placebo trials did not differ in their respective absolute hypocapnic reactivities. Relative hypocapnic QICA and MCAv reactivity were reduced by 60±32% (P<0.001) and 76±32% (P<0.001), respectively, following INDO, but were unaltered following the placebo treatment. For both QICA and MCAv, the pre-INDO, pre-placebo, and placebo trials did not differ in their respective relative hypocapnic reactivities.   Figure 2.0.  Volumetric flow and velocity cerebrovascular reactivity to hypercapnia. A. Hypercapnic QICA reactivity prior to and following INDO and placebo treatments. B. Hypercapnic MCAv reactivity prior to and following INDO and placebo treatments. Horizontal lines denote significant differences between trials, P<0.001.   Hypercapnic reactivity was greater than that during hypocapnia for QICA and MCAv during all trials. During hypercapnia, QICA reactivity was greater than that of both ICA velocity (ICAv; 7.3±2.0 vs. 3.5±2.2 % ⋅ mmHgPETCO2; P<0.001) and MCAv (7.3±2.0 vs. 4.8±0.9 % ⋅ mmHgPETCO2; P<0.001) reactivity, which could be explained by progressive dilation of the ICA (see “2.4.3. Vasomotor response to CO2”). There was no difference in hypocapnic QICA, ICAv, and MCAv reactivity, likely due to a modest vasomotor reactivity to hypocapnia.   ! 56! Figure 2.1. Volumetric flow and velocity cerebrovascular reactivity to hypocapnia. A. Hypocapnic QICA reactivity prior to and following INDO and placebo treatments. B. Hypocapnic MCAv reactivity prior to and following INDO and placebo treatments. Horizontal lines denote significant differences between trials, P<0.001.   2.4.3. Vasomotor response of the internal carotid artery to CO2.   Prior to INDO, the ICA dilated significantly at +6 & +9 mmHg PETCO2; although to a lesser extent, dilation still occurred at +6 & +9mmHgPETCO2 following INDO (Figure 2.2-A). The ICA dilated at every stage of hypercapnia pre and post placebo treatment (Figure 2.2-B.). During hypocapnia the ICA constricted at -9mmHgPETCO2 prior to INDO, with no constriction post-INDO (Figure 2.3-A). During both the pre-placebo and placebo trial, the ICA constricted at -6 & -9mmHg PETCO2 (Figure 2.3-B).      ! 57! Figure 2.2. The vasomotor response to hypercapnia. A. The vasomotor response to hypercapnia pre () and post (○) INDO. B. The vasomotor response to hypercapnia pre (■) and post (□) placebo. * difference from baseline P<0.05; ** difference from baseline P<0.01; *** difference from baseline P<0.001; # difference from previous stage P<0.05; ## difference from previous stage P<0.01; ### difference from previous stage P<0.001; † within day difference in diameter between baseline and intervention P<0.05; †† within day difference in diameter between baseline and intervention P<0.01; ††† within day difference in diameter between baseline and intervention P<0.001.   The slope of the vasomotor response to hypercapnia was reduced by 67±28% following INDO (0.045±0.015 vs. 0.015±0.012 mm ⋅ mmHgPETCO2-1; P<0.001) but was unaffected by placebo (0.036±0.006 vs. 0.033±0.006 mm ⋅ mmHgPETCO2-1; P=0.25). No change in the slope of the response to hypocapnia occurred following INDO (0.019±0.015 vs. 0.006±0.008 mm ⋅ mmHgPETCO2-1; P=0.08). Vasomotor reactivity was greater during hypercapnia than during hypocapnia (0.045±0.015 vs. 0.019±0.015 mm ⋅ mmHgPETCO2-1; P<0.01).     ! 58! Figure 2.3. The vasomotor response to hypocapnia. A. The vasomotor response to hypocapnia pre () and post () INDO. B. The vasomotor response to hypocapnia pre () and post (□) placebo. * difference from baseline P<0.05; ** difference from baseline P<0.01; *** difference from baseline P<0.001; † within day difference in diameter between baseline and intervention P<0.05; †† within day difference in diameter between baseline and intervention P<0.01; ††† within day difference in diameter between baseline and intervention P<0.001.    2.4.4. Effect of Ketorolac on CO2 mediated responses (Table 2.1).   Administration of Ketorolac had no effect on resting ventilation (11.3±1.5 vs. 11.8±1.9 L ⋅ min-1; P=0.65), MAP (79.2±8.6 vs. 80.0±5.9 mmHg; P=0.79), HR (55.1±12.5 VS. 53.9±14.6 bpm; P=0.54), QICA (282.8±66.7 vs. 295.6±80.3 mL ⋅ min-1; P=0.38), or MCAv (56.5± 13.3 vs. 58.3±8.1 cm ⋅ s-1; P=0.54). Following Ketorolac there was no change in absolute QICA or MCAv reactivity during both hypercapnia and hypocapnia. Relative reactivities for QICA and MCAv did not differ before and after Ketorolac either. The vasomotor response of the ICA to both hypercapnia (0.026±0.015 vs. 0.026±0.022 mm ⋅ mmHgPETCO2-1; P=0.99) and hypocapnia (0.020±0.024 vs. 0.017±0.014 mm ⋅ mmHgPETCO2-1; P=0.73) were unchanged following Ketorolac.       ! 59!Table 2.1. Cerebrovascular and blood pressure responses to CO2 before and following Ketorolac.  Hypercapnia   Absolute Reactivity  Relative Reactivity  Units Pre-Ketorolac Ketorolac Units Pre-Ketorolac Ketorolac QICA (mL · min-1 · mmHgPETCO2-1) 16.6±8.6+ 16.2±8.5+ (% · mmHgPETCO2-1) 5.5±2.5 5.2±2.0+ MCAv (cm · s-1 · mmHgPETCO2-1) 2.6±0.6‡ 2.7±0.6‡ (% · mmHgPETCO2-1) 4.6±0.7† 4.6±1.2† VE (L · min-1) 3.0±2.0 2.8±1.74 (% · min-1) 23.7±13.9 19.0±9.6 MAP (mmHg · mmHgPETCO2-1) 1.1±0.7† 0.8±0.5+ (% · mmHgPETCO2-1) 1.4±0.9† 1.0±0.5+  Hypocapnia   Absolute Reactivity  Relative Reactivity  Units Pre-Ketorolac Ketorolac Units Pre-Ketorolac Ketorolac QICA (mL · min-1 · mmHgPETCO2-1) 8.0±2.1 6.5±4.0 (% · mmHgPETCO2-1) 2.8±0.7 2.2±0.9 MCAv (cm · s-1 · mmHgPETCO2-1) 1.2±0.6 1.7±0.7 (% · mmHgPETCO2-1) 2.1±0.8 2.9±1.4 MAP (mmHg · mmHgPETCO2-1) -0.4±0.3 -0.2±0.4 (% · mmHgPETCO2-1) -0.5±0.3 -0.3±0.5 + significant difference between hypercapnic and hypocapnic reactivities within an experimental trial, P<0.05; † significant difference between hypercapnic and hypocapnic reactivities within an experimental trial, P<0.01; ‡ significant difference between hypercapnic and hypocapnic reactivities within an experimental trial, P<0.001.   2.5. Discussion.  The main novel findings of this study were: 1) The ICA dilates in response to hypercapnia and modestly constricts in response to hypocapnia; 2) Vasomotion of the ICA in response to hypercapnia is markedly blunted (-67%) following INDO administration; and 3) INDO, but not Ketorolac, inhibits cerebrovascular responses to CO2, indicating a permissive (i.e., non-COX inhibition mediated) effect(s) of INDO.   2.5.1. Cerebrovascular responses to CO2.   Our volumetric data have revealed similar hypercapnic reactivity values to that first collected using the Kety & Schmidt technique (Kety & Schmidt, 1948b) of ~7-8 % ⋅ mmHgPETCO2-1; such values are consistent with recent studies (Willie et al., 2012; Hoiland et al., 2015). This ! 60!volumetric reactivity, during hypercapnia, greatly exceeds that recorded using TCD measures of MCAv in the current and previous studies [e.g., (Ide et al., 2003, 2007; Willie et al., 2012; Regan et al., 2014; Brothers et al., 2014; Hoiland et al., 2015)]. Since MCAv and ICAv reactivity to hypercapnia did not differ, this indicates that vasomotion is responsible for the difference between volumetric flow and velocity indices of cerebrovascular reactivity. Consistent with this finding, are recent reports that the MCA dilates in response to hypercapnia (Verbree et al., 2014; Coverdale et al., 2014, 2015) leading to consequent underestimation of cerebrovascular reactivity with velocity measures (Ainslie & Hoiland, 2014).   The vasomotor response of the ICA to hypercapnia was markedly greater than that during hypocapnia; thus, there was no appreciable difference in volumetric and velocity indices of reactivity in the hypocapnic range. Upon close examination of the vasomotor profile of the ICA during changes in PETCO2, it is quite similar to the expected vasomotor profile of the MCA diameter (see Figure 1 in: Ainslie & Hoiland, 2014), wherein small changes in PETCO2 do not elicit measurable changes in diameter. It seems that a larger hypocapnic stimulus (e.g.,  ~9mmHg below baseline) is needed to elicit constriction than the necessary hypercapnic stimulus for dilation, likely a function of the overall lower cerebrovascular reactivity during hypocapnia. A mechanistic basis for these differences is not currently known. However, using our experimental approach (see below “2.5.2. comparison between studies” for explanation) it appears that the ICA may be more sensitive to changes in CO2 in the hypercapnic range as dilation at +3mmHgPETCO2 was observed in both the pre- and post-placebo trials, while the MCA does not appear to dilate (as measured using 3-T MRI) until approximately +9mmHgPETCO2 (Coverdale et al., 2014).   2.5.2. Comparisons between studies.  Conflicting data exists as to whether CO2 mediated vasomotion of the ICA does (Willie et al., 2012; Hoiland et al., 2014) or does not occur (Coverdale et al., 2015). The present study aimed to resolve these conflicting data, and corroborates the data exemplifying that vasomotion of the ICA does occur. Differences between studies are likely due to several ! 61!methodological and analytical differences. These differences include the method of manipulating PaCO2, and determination of vessel diameter. For example, Coverdale et al., 2015 utilized elevations in FiCO2 (0.06) to elicit a ~10mmHg increase in PETCO2. This technique is limited in that PETCO2 tends to overestimate PaCO2 by ~4-6mmHg at this level of hypercapnia (Peebles et al., 2007). In contrast, our approach of end-tidal forcing, only results in PETCO2 over estimating PaCO2 by ~1mmHg (Tymko et al., 2015). Therefore, it is likely that the hypercapnic PaCO2 (not PETCO2) stimulus applied by Coverdale et al., 2015 was only around 5mmHg above baseline whereas the PETCO2 stimulus in the current study is representative of the true PaCO2 stimulus. As the current study shows very modest changes in diameter with mild hypercapnia, this provides an explanation, in part, for the lack of ICA dilation observed by Coverdale et al., 2015. Second, and importantly, discrepancies between studies may be a reflection of the use of caliper based manual diameter measures (quantified over three cardiac cycles) rather than our automated approach (quantified over a minimum of 12 consecutive cardiac cycles). The latter approach is affected less by artifacts, and the potential influences of respiration and blood pressure variability. As such, using edge-detection software not only provides a more robust and sensitive assessment of vessels diameter and velocity (Woodman et al., 2001), it also limits subjectivity and bias during data analysis.   2.5.3. Cyclooxyygenase inhibition.   It has been previously established that the dose of INDO in the current study is sufficient to effectively inhibit prostaglandin synthesis (Eriksson et al., 1983). Inhibition of COX with INDO resulted in a significant reduction in both resting CBF and cerebrovascular CO2 reactivity in both the hypo- and hyper-capnic range. This result is in concordance with previous studies using various measurement techniques to quantify CBF (Wennmalm et al., 1981; Bruhn et al., 2001; St Lawrence et al., 2002; Xie et al., 2006; Hoiland et al., 2015). However, COX inhibition with Ketorolac did not affect resting CBF, cerebrovascular CO2 reactivity, or the vasomotor response of the ICA. Some previous data, comparing INDO to other COX inhibitors, has indicated that INDO reduces CBF and cerebrovascular CO2 reactivity via a ‘COX inhibition independent’ mechanism (Eriksson et al., 1983; Markus et ! 62!al., 1994); however, these studies provided limited suggestion of alternative mechanisms. On the basis of previous studies in vitro and in highly controlled animal models, we reason here that the selectivity of INDO is related to either the inhibition of cAMP-dependent protein kinase (Kantor & Hampton, 1978), inhibition of prostaglandin receptor binding mediated increases in vascular smooth muscle cAMP (Parfenova et al., 1995b), or both (see “2.5.4. Pharmacological interventions” below). 2.5.4. Pharmacological interventions.   Several factors have supported INDO as a useful pharmacological agent to assess the effects of COX inhibition on CBF:  cardiac output is minimally affected by the administration of INDO (Nowak & Wennmalm, 1978; Wennmalm et al., 1984) and INDO  does not appear to alter cerebral metabolism (Kraaier et al., 1992), resting ventilation (Hoiland et al., 2015), peripheral chemosensitivity (Xie et al., 2006), or plasma catecholamines (Staessen et al., 1984; Green et al., 1987). Thus, INDO has been extensively used to assess the influence of prostaglandins on cerebrovascular CO2 reactivity (Wennmalm et al., 1981; Eriksson et al., 1983; Bruhn et al., 2001; St Lawrence et al., 2002) in addition to many other experimental paradigms. The assumption was made that INDO reduces cerebrovascular CO2 reactivity through a prostaglandin independent mechanism, a conclusion drawn from divergent findings using other COX inhibiting drugs such as Aspirin and Naproxen (Eriksson et al., 1983; Markus et al., 1994). Yet, it is still continually used to assess prostaglandin-mediated responses. For example, extremely low doses of INDO have been reported to inhibit cAMP – dependent protein kinase activity (Kantor & Hampton, 1978). This inhibition will directly effect vascular smooth muscle cell calcium sensitivity (Adelstein & Conti, 1978), and thus vasomotor tone (Kerrick & Hoar, 1981). Moreover, in newborn pigs, which also demonstrate reductions in CBF/CO2 reactivity to INDO but not other COX inhibitors (e.g., Aspirin, Ibuprofen, Naproxen; Chemtob et al., 1991), INDO blocks prostaglandin receptor mediated signal transduction during hypercapnia (Parfenova et al., 1995b), specifically downstream increases in cAMP. However, Aspirin does not inhibit prostaglandin receptor mediated signalling and subsequent increases in smooth muscle cell cAMP (Parfenova et al., 1995b), highlighting that this is a unique effect of INDO.  Therefore, although the primary purpose of ! 63!this study was to characterize the effect of INDO on CO2 mediated vasomotor responses of the ICA, we also aimed to test the effect of Ketorolac on such vasomotor responses. This agent was to determine if INDO acts differently from other COX inhibitors in its ability to blunt CO2 mediated vasomotion (in addition to flow reactivity) of large extra-cranial cerebral arteries. With Ketorolac we showed no difference in the vasomotor response, consistent with these aforementioned studies (Eriksson et al., 1983; Markus et al., 1994), indicating that INDO is also affecting the vasomotor tone of larger cerebral arteries in a permissive (i.e., independent of COX inhibition) manner.  The possibility remains that only low levels of prostaglandins are required to induce vasomotion (through downstream increases in cAMP) during CO2 perturbations. This possibility provides an explanation for the lack of effect of COX inhibitors - with the exception of INDO - on cerebrovascular CO2 reactivity, consistent with the lack of detectable levels of prostaglandins during hypercapnia (Eriksson et al., 1983). For example, near complete inhibition of prostaglandin synthesis with Aspirin does not inhibit prostaglandin receptor agonism (i.e., via Iloprost) or hypercapnia mediated increases in cAMP and vascular diameter in newborn pigs (Parfenova et al., 1995b). If large doses of prostaglandins were necessary to produce dilation one may expect a dose-dependent relationship between prostaglandin receptor agonism and vasodilation; however, this is not the case during hypercapnia (Leffler et al., 1999). Collectively, these data provide a possible explanation (i.e., inhibition of prostaglandin receptor mediated signal transduction) as to how INDO inhibits cerebrovascular CO2 reactivity both independent of COX inhibition and in a manner unique from other COX inhibitors.   2.5.5. Implications.   For the last 30 years, assessment of cerebrovascular responses has been dominated by the use of TCD. However, assessment of CO2 reactivity with TCD operates on the assumption (also its primary limitation) that the diameter of the MCA does not change in response to CO2 - an assumption previously thought to be true (Serrador et al., 2000). Recently, studies using higher resolution magnetic resonance imaging (Verbree et al., 2014; Coverdale et al., 2014) ! 64!have reported MCA vasomotion in response to changes in CO2 and consequent underestimation of reactivity by TCD measures of velocity. Other studies assessing volumetric flow reactivity through the extra-cranial cerebral arteries (Willie et al., 2012; Hoiland et al., 2015) provide further evidence that TCD is limited in its ability to accurately measure cerebrovascular CO2 reactivity.   Reduced cerebrovascular CO2 reactivity is indicative of an increased risk for all cause and cardiovascular (inclusive of stroke) mortality (Portegies et al., 2014). Since it seems that changes in diameter can contribute to nearly half of the increase in flow observed during elevations in PETCO2, we speculate that the magnitude of the vasomotor response (i.e., diameter change) in response to PETCO2 perturbations may be indicative of cerebrovascular health (i.e., endothelial function), much like peripheral flow mediated dilation is indicative of cardiovascular risk (Inaba et al., 2010). The incorporation of diameter measures into the prediction of cerebrovascular events and related mortality is now needed.   While INDO is exemplary in its ability to reduce cerebrovascular reactivity, and is thus an attractive tool for the assessment of physiological function associated with cerebrovascular reactivity (i.e., control of breathing; Xie et al., 2006; Hoiland et al., 2015), its utility for investigating the role of prostaglandins in mediating cerebrovascular responses should be cautioned. As we and others (Eriksson et al., 1983; Markus et al., 1994) have identified, INDO reduces cerebrovascular CO2 reactivity and CO2 mediated vasomotion in a manner that is unique from other COX inhibitors. It is clear that INDO is acting in a prostaglandin independent manner, likely through the inhibition of cAMP-dependent protein kinase (Kantor & Hampton, 1978), and reductions in prostaglandin receptor mediated increases in smooth muscle cAMP (Parfenova et al., 1995b) to affect cerebrovascular reactivity to CO2.  2.6. Synopsis.  We have demonstrated for the first time that INDO reduces the vasomotor response of the ICA to changes in PETCO2 and provided evidence that it is independent of inhibiting COX. Future studies should aim to determine the mechanism(s) underlying INDO’s unique ability ! 65!to reduce CO2 mediated cerebrovascular vasomotion and to determine other regulatory mechanisms governing cerebrovascular vasomotion in healthy humans.  2.7. Author Contributions.  Conception and design of experiments: RLH, PNA. Data Collection: RLH, MMT, KWW, ARB, BM, PNA. Data analysis and interpretation: RLH, PNA. Manuscript first draft: RLH, PNA. Critical revisions of manuscript for important intellectual content: RLH, MMT, KWW, ARB, BM, PNA. Approval of final draft: RLH, PNA.  2.8. Funding.  This research was supported by an NSERC Discovery Grant and Canadian Research Chair in Cerebrovascular Physiology (PNA).  2.9. Special Recognition.   Special thanks to Dr. Glen Foster for aiding in the optimization and the functioning of the end-tidal forcing system and technical support.     ! 66!Chapter 3. Conclusion. !3.1. Indomethacin induced impairments of cerebrovascular reactivity.  It has been well documented that INDO reduces cerebrovascular CO2 reactivity by approximately 40-60% (Eriksson et al., 1983; Xie et al., 2006; Hoiland et al., 2015). However, it is not as well documented if INDO inhibits cerebrovascular reactivity in a regional or sex dependent manner, and if chronic treatment has a differential effect on reactivity from that of acute treatment. There is some evidence that INDO inhibits the reactivity of grey matter to a greater extent than white matter (St Lawrence et al., 2002); however, this may simply be a product of grey matter possessing a much higher reactivity than white matter in normal conditions (Ramsay et al., 1993), and thus a greater potential for reductions in reactivity with INDO administration.  For the influence of sex on INDO mediated inhibition of cerebrovascular reactivity, there is evidence suggesting that INDO impairs cerebrovascular CO2 reactivity to a greater extent in females than in males (Kastrup et al., 1999). Of note, this study utilized TCD impairing our ability to firmly conclude sex differences are present (see “1.4.1. Transcranial Doppler ultrasound” for limitations of TCD). Further, as it has been clearly outlined already, due to INDO’s unique efficacy for reducing cerebrovascular CO2 reactivity (Eriksson et al., 1983), the reported sex differences are likely unrelated to COX. As noted by Kastrup et al., 1999, the larger reduction seen in females is likely due to a higher reactivity prior to INDO and resultant greater potential for larger reductions in reactivity to occur. Indeed, the authors noted that the magnitude of reductions in cerebrovascular CO2 reactivity were positively correlated with resting MCAv (r=0.74). However, data collected in rats support an interaction between estrogen levels and the vasodilatory and constrictive balance of prostanoids. For example, estrogen increases cerebral prostacyclin production in rats (Ospina et al., 2003) shifting the balance towards more vasodilatory prostanoids indacting sex differences may be important. This raises the possibility that inhibition of COX with INDO likely produces larger reductions in cerebrovascular CO2 reactivity in biological systems with higher estrogen levels. Further study is required to determine if the effects of ! 67!prostaglandin inhibition differ between males and females in humans. In the present study, the single female subject was tested on days 1 and 3 of her follicular phase, when estrogen levels are the lowest relative to the rest of the menstrual cycle (Marsh et al., 2011).   Related to the chronic use of INDO, there is little and conflicting data as to the effects on cerebrovascular CO2 reactivity. A study by Eriksson et al., 1983, using the Kety & Schmidt technique (Kety & Schmidt, 1948a) to assess CBF,  reported that after 1-week of chronic INDO (0.8mg/Kg three times daily), cerebrovascular CO2 reactivity was reduced to the same extent as that observed after acute administration of INDO. Reactivity after chronic INDO administration was tested 2-3 hours after the final dose of INDO (Eriksson et al., 1983). In contrast to these findings, Pickles et al., (1984) reported that after three days of treatment with INDO (100mg/day in divided doses), cerebrovascular CO2 reactivity was not different from that assessed prior to INDO (Pickles et al., 1984). In this study, using the Xenon133 clearance technique to assess CBF it was not reported how long after the final dose of INDO reactivity was assessed. These contrasting studies make difficult the interpretation of the effect of chronic INDO treatment on cerebrovascular CO2 reactivity and, therefore, warrant further study of this topic. If chronic INDO dosing does reduce cerebrovascular CO2 reactivity, especially in the elderly, this reduction in reactivity may be pre-disposing people to the risk of cardiovascular/cerebrovascular events to an extent greater than that assumed as a result of drug induced changes in the pro- versus anti-thrombotic prostanoid balance.  3.2. Cerebrovascular CO2 reactivity: Implications for disease.  3.2.1. Sleep apnea.  Cerebrovascular reactivity is an integral component in the control of breathing both in an awake (Fan et al., 2010) and sleeping (Xie et al., 2009) state, with pharmacological reductions in reactivity leading to breathing instability. Accordingly, a blunted cerebrovascular CO2 reactivity is fundamental to the pathogenesis of central (Javaheri & Dempsey, 2013) and obstructive (Dempsey et al., 2010) sleep apnea, diseases both characterized by ventilatory dysregulation. For example, INDO induced reductions in ! 68!cerebrovascular CO2 reactivity have been reported to worsen obstructive sleep apnea at sea level (Burgess et al., 2010), and central sleep apnea at high-altitude (Burgess et al., 2014). Blunted reactivity is also implicated in the pathogenesis of central sleep apnea in heart failure patients (Xie et al., 2005). Therefore, treatment of central and obstructive sleep apnea may benefit from an improved understanding of the underlying mechanisms regulating cerebrovascular CO2 reactivity and the development of therapeutic strategies to increase cerebrovascular CO2 reactivity (i.e., exercise training in the elderly; Ainslie et al., 2008).    3.2.2. Prediction of mortality.   Assessment of cerebrovascular CO2 reactivity in a general population has been shown as an effective modality for the prediction of all cause and cardiovascular mortality (Portegies et al., 2014). The study by Portegies et al., 2014 reported that reduced cerebrovascular CO2 reactivity was not associated with incidence of stroke despite its relation to both cardiovascular and all cause mortality. While these findings may be interpreted as discrediting to the value of cerebrovascular CO2 reactivity as a predictor of cerebrovascular disease, they may simply be limited in that impaired vascular function can manifest in a variety of pathological conditions in addition to stroke, especially if vascular function is systemically impaired. Moreover, as this study utilized TCD, and cerebral vasomotion contributes largely to reactivity, between-individual variations in the dilatory response to hypercapnia will likely have affected the velocity reactivity profiles in an undetectable manner and potentially led to miss-classification of subjects. It is plausible that incorporation of the dilatory response to hypercapnia into such a study would improve the ability to predict cerebrovascular and cardiovascular events. Whilst the efficacy of cerebrovascular CO2 reactivity in the prediction of cerebrovascular events in otherwise healthy subject requires much further study, the link between reduced reactivity and risk of stroke is much more robust in patients already suffering from cerebrovascular disease such as carotid stenosis [e.g. (Gupta et al., 2012)]. Therefore, while pre-existing cerebrovascular disease clearly predisposes individuals to an increased risk of stroke with reduced cerebrovascular CO2 reactivity, furthering the understanding of the link between cerebrovascular reactivity and risk for stroke remains an important task (Bos et al., 2007).  ! 69! Currently, the mechanisms underlying reduced cerebrovascular reactivity relative to a normal healthy aging population, as was used by Portegies et al., 2014 (~70 years of age), are relatively unknown. There are data, however, attributing reduced prostaglandin-mediated dilation to the overall reduction of cerebrovascular CO2 reactivity associated with aging (Barnes et al., 2012). This indicates a potential role for a loss of vasodilatory prostaglandins (i.e., prostacyclin) being responsible, at least in part, for reductions in cerebrovascular CO2 reactivity associated with vascular disease and increased risk of mortality. Moreover, a shift in the balance between anti-thrombotic (i.e., prostacyclin) and pro-thrombotic (i.e., thromboxane) prostanoids is likely also responsible (via shifting to a pro-thrombotic state). It is important to note that the effect of INDO observed by Barnes et al., 2012 is likely downstream of COX inhibition and manifesting through inhibition of PGI2 and PGE2 receptor mediated signal transduction and related modulation of smooth muscle cell cAMP (Parfenova et al., 1995b).    3.3. Methodological limitations.  Two different non-selective COX inhibitors were used in this study to investigate the role of PG’s in mediating cerebrovascular CO2 reactivity and the vasomotor response of the ICA to changes in PaCO2. For this purpose, end-tidal forcing of PETCO2 was used as a modality for manipulating PaCO2 and driving changes in CBF. However, both of these methods, the pharmacological inhibition of COX and the use of PETCO2 as a non-invasive surrogate of PaCO2, come with inherent limitations. These limitations will be discussed below.  3.3.1. Pharmacological inhibition of cyclooxygenase.    In the present study we used either 1.2mg/Kg INDO or 0.25mg/Kg Ketorolac to inhibit COX activity and consequently the conversion of arachidonic acid to PGH2. Such measurement of the efficacy of COX inhibition requires the sampling of venous blood, centrifuging of the blood to extract platelet rich plasma samples, and subsequent addition of Na-arachidonate to the plasma sample to assess platelet aggregation. For example, Eriksson et al., 1983 showed ! 70!that prior to Aspirin, addition of Na-arachidonate to a platelet rich plasma sample lead to platelet aggregation, whereas, following treatment with INDO, platelet aggregation in response to Na-arachidonate was completely inhibited signifying prostaglandin synthesis was abolished (Eriksson et al., 1983). While we did not assess the efficacy of our INDO administration in inhibiting prostaglandin synthesis, we used a similar dosage as that shown to be effective by Eriksson et al., 1983.  As INDO possesses effects additional to the inhibition of COX (i.e., inhibition of cAMP-dependent protein kinase) we chose to also test an alternate non-selective COX inhibitor, Ketorolac. By using Ketorolac we were able to assess the role of COX activity in mediating cerebral vasomotor responses to changes in PETCO2 without the confounder of simultaneous cAMP-dependent protein kinase inhibition. As no effect of Ketorolac was observed on the vasomotor response to hypercapnia and hypocapnia, the obvious conclusion would be that prostaglandins are not an obligatory mediator of cerebral vasomotor responses to changes in PETCO2. However, there is evidence in animals, that downstream from COX inhibition, prostaglandin receptor mediated increases in cAMP are responsible for dilation of cerebral vessels during hypercapnia (Parfenova et al., 1995b). Specifically, inhibition of COX with Aspirin has no effect on IP receptor (prostacyclin receptor) agonism or hypercapnia mediated increases in cAMP and subsequent vessel dilation, but INDO reduces both IP receptor agonism and hypercapnia mediated increases in cAMP and subsequent vessel dilation (Parfenova et al., 1995b). These findings by Parfenova et al., 1995 would indicate that despite COX inhibition in the Aspirin trial, prostaglandins are still mediating dilation, whereas, INDO inhibits IP receptor specific increases in smooth muscle cell cAMP that are essential to cerebral vessel dilation during hypercapnia. This provides evidence for the theory that only low levels of prostaglandins (too low to be affected by COX inhibition) are required for cerebral vessel dilation to hypercapnia – in support of this,  prostacyclin does not have a dose dependent effect on facilitating hypercapnic dilation (Parfenova & Leffler, 1996), and CBF increases in response to hypercapnia despite no observed increases in prostaglandin production in humans (Eriksson et al., 1983). Exactly how INDO inhibits IP receptor mediated increases in smooth muscle cell cAMP and vessel dilation is not currently ! 71!known, and requires further investigation in highly controlled animal models prior to translation into human studies.  3.3.2. End-tidal versus arterial CO2.   In the current study we used end-tidal forcing to control PETCO2, a commonly used surrogate for PaCO2. While various methods can be used to manipulate PETCO2 for the purpose of measuring cerebrovascular responses (see “1.5. Experimental models to assess cerebrovascular CO2 reactivity”), we chose to use end-tidal forcing as it minimizes the gradient between PETCO2 and PaCO2 (~2mmHg; Tymko et al., 2015), while allowing for the assessment of CBF during steady state conditions. Previously collected data shows that during end-tidal forcing the PETCO2-PaCO2 gradient does not change from baseline to hypercapnia or hypocapnia (Tymko et al., 2015); therefore, it is evident that our PETCO2 stimulus magnitude is representative of the PaCO2 stimulus magnitude. Moreover, while ventilatory sensitivity will affect the necessary FiCO2 of the inspirate and subsequently the level of CO2 rebreathed from apparatus deadspace (Tymko et al., 2015), ventilatory sensitivity to CO2 was not different between trials for either drug intervention. As such, ventilation will not have impacted on the PETCO2 and PaCO2 gradient between trials.  3.4. Future studies.  This study has indicated that vasodilation of the ICA is an integral component to the increases in flow observed during hypercapnia. Moreover, these data support an effect of INDO on cerebrovascular CO2 reactivity that is independent of COX inhibition. Thus, future studies should endeavor to further the understanding of the cellular mechanisms regulating cerebrovascular CO2 reactivity, the mechanisms behind INDO’s unique ability to blunt cerebrovascular CO2 reactivity compared to other COX inhibitors, and the implications of large extra-cranial cerebral vasomotion in the context of aging and disease.   ! 72!3.4.1. Determining the role of cyclic adenosine monophosphate inhibition on cerebrovascular vasomotor responses in humans.   As INDO inhibits cAMP-dependent protein kinase at submicromolar doses, and is ~80-220 times more potent an inhibitor of cAMP-dependent kinase than it is of either COX isoform, one could theoretically use low doses of this drug to determine the role of cAMP in regulating cerebrovascular responses independent of COX. The potential use of this approach would require pilot work to assess the necessary oral dose of INDO needed to reduce cerebrovascular vasomotor response to CO2, while prostaglandin synthesis remains intact. Using previously established methods (Eriksson et al., 1983), the extent of prostaglandin synthesis inhibition for various doses of INDO could then be established. Following establishment of this effect in several subjects, the desired dose could be administered to a larger group of subjects (i.e., n≈10) to assess if cerebrovascular vasomotor responsiveness is reduced despite prostaglandin synthesis remaining intact.   3.4.2. The assessment of cerebral endothelial function.   Endothelial function of the peripheral vasculature, assessed by flow mediated dilation, is predictive of cardiovascular events (Inaba et al., 2010). If an assessment of cerebral endothelial function were available it would have the potential to be an effective predictor of cerebrovascular events. Thus, a test to determine the flow-mediated dilation of cerebral vessels is needed. As the ICA is bilaterally responsible for supplying ~70% of the brains blood supply and can be insonated in most individuals, it provides an attractive avenue to assess cerebrovascular responses. Future study should assess the applicability of a shear stress test of the ICA, similar to that done in the brachial artery to assess peripheral endothelial function. If highly controlled and short term increases in PETCO2 are achievable with end-tidal forcing, PETCO2 may provide an avenue to manipulate shear stress in a time course similar to that induced by cuff release in brachial tests. For example, abrupt (30 second) increases in CO2, with a rapid return to baseline will result in a large (~3-fold) increase in shear rate (Hoiland et al., unpublished findings) mimicking the shear profile following brachial cuff release. By rapidly returning PETCO2 to baseline it is likely that any ! 73!any vascular changes are occurring as a result of the shear stimulus. The extent to which these changes in shear patterns may provide insight into the endothelial function of the cerebral vasculature remains to be fully explored.  3.4.3. Vasomotor responsiveness of the internal carotid artery with aging.  It has been previously established that CO2 reactivity is reduced with aging, likely due to a reduction in vasodilatory prostaglandin production (Barnes et al., 2012). Thus, future studies should seek to determine whether or not the vasodilatory response of the ICA is impaired with age. Such a study would require replication of the CO2 reactivity tests used in this thesis and concurrent measuring of QICA and vasomotor reactivity to CO2 in young and old subjects. Further, it should be determined if COX inhibition affects CO2 reactivity in older individuals following chronic usage of COX inhibitors. Such chronic use of COX inhibitors (typically 7-10 days) is common in elderly populations despite the well-know increased risk of cardiovascular events from such medications (Brune & Patrignani, 2015).  3.4.4. Other regulatory factors in the cerebral vasomotor response to CO2 perturbations.  As outlined in the introduction of this thesis (see “1.2.3 Potential cellular mechanisms mediating cerebrovascular CO2 reactivity”), several vasoactive factors have been postulated to contribute to the cerebral vasomotor response to CO2. However, the vast majority of these data have been collected in animal models, with related studies in human lacking. Thus, the present study provides a well-controlled model (i.e., ICA vasomotion) that can be used to explore the effects of other pharmacological interventions on the human cerebral vasomotor response to CO2. For example, the current study could be repeated using theophylline (adenosine receptor antagonists), Nω-nitro-L-arginine methyl ester (NOS inhibitor), or glibenclamide (an KATP channel inhibitor). Much research is needed to better understand the fundamental mechanisms of cerebrovascular regulation and potential interactions between vasoactive factors.     ! 74!Bibliography Aaslid R, Markwalder TM & Nornes H (1982). Noninvasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 57, 769–774. Abdel-halim MS, Holst H Von, Meyerson B, Sachs C & Anggard E (1980). Prostaglandin Profiles in Tissue and Blood Vessels from Human Brain. J Neurochem 34, 1331–1333. Adelstein R & Conti M (1978). Phosphorylation of Smooth Muscle Myosin Catalytic subunit of adenosine 3’: 5'-monophosphate-dependent protein kinase. J Biol Chem 253, 8347–8350. Ainslie PN, Ashmead JC, Ide K, Morgan BJ & Poulin MJ (2005). Differential responses to CO2 and sympathetic stimulation in the cerebral and femoral circulations in humans. J Physiol 566, 613–624. Ainslie PN, Cotter JD, George KP, Lucas S, Murrell C, Shave R, Thomas KN, Williams MJ a & Atkinson G (2008). Elevation in cerebral blood flow velocity with aerobic fitness throughout healthy human ageing. J Physiol 586, 4005–4010. Ainslie PN & Duffin J (2009). Integration of cerebrovascular CO2 reactivity and chemoreflex control of breathing: mechanisms of regulation, measurement, and interpretation. AJP Regul Integr Comp Physiol 296, R1473–R1495. Ainslie PN & Hoiland RL (2014). Transcranial Doppler Ultrasound: Valid, Invalid, or Both? J Appl Physiol 117, 1081–1083. Ainslie PN, Murrell C, Peebles K, Swart M, Skinner MA, Williams MJA & Taylor RD (2007). Early morning impairment in cerebral autoregulation and cerebrovascular CO2 reactivity in healthy humans: relation to endothelial function. Exp Physiol 92, 769–777. Ainslie PN & Poulin MJ (2004). Ventilatory, cerebrovascular, and cardiovascular interactions in acute hypoxia: regulation by carbon dioxide. J Appl Physiol 97, 149–159. Ainslie PN, Shaw AD, Smith KJ, Willie CK, Ikeda K, Graham J & Macleod DB (2014). Stability of cerebral metabolism and substrate availability in humans during hypoxia and hyperoxia. Clin Sci (Lond) 126, 661–670. Ainslie PN & Subudhi AW (2014). Cerebral blood flow at high altitude. High Alt Med Biol 15, 133–140. Attwell D, Buchan AM, Charpak S, Lauritzen M, Macvicar BA & Newman EA (2010). Glial and neuronal control of brain blood flow. Nature 468, 232–243. Attwell D & Laughlin SB (2001). An energy budget for signaling in the grey matter of the brain. J Cereb blood flow Metab 21, 1133–1145. ! 75!Barnes JN, Schmidt JE, Nicholson WT & Joyner MJ (2012). Cyclooxygenase inhibition abolishes age-related differences in cerebral vasodilator responses to hypercapnia. J Appl Physiol 112, 1884–1890. Bayerle-Eder M, Wolzt M, Polska E, Langenberger H, Pleiner J, Teherani D, Rainer G, Polak K, Eichler HG & Schmetterer L (2000). Hypercapnia-induced cerebral and ocular vasodilation is not altered by glibenclamide in humans. Am J Physiol Regul Integr Comp Physiol 278, 1667–1673. Black MA, Cable NT, Thijssen DHJ & Green DJ (2008). Importance of measuring the time course of flow-mediated dilatation in humans. Hypertension 51, 203–210. Blaha M, Benes V, Douville CM & Newell DW (2007). The effect of caffeine on dilated cerebral circulation and on diagnostic CO2 reactivity testing. J Clin Neurosci 14, 464–467. Bleys RL, Cowen T, Groen GJ, Hillen B & Ibrahim NB (1996). Perivascular Nerves of the Human Basal Cerebral Arteries: I. Topographical Distribution. J Cereb Blood Flow Metab 16, 1034–1047. Bos MJ, Koudstaal PJ, Hofman A, Witteman JCM & Breteler MMB (2007). Transcranial Doppler hemodynamic parameters and risk of stroke: the Rotterdam study. Stroke 38, 2453–2458. Boullin D, Bunting S, Blaso WP, Hunt TM & Moncada S (1979). Responses of human and baboon arteries to prostaglandin endoperoxides and biologically generated and synthetic prostacyclin: their relevance to cerebral arterial spasm in man. Br J Pharmacol139–147. Brothers RM, Lucas RAI, Zhu Y-S, Crandall CG & Zhang R (2014). Cerebral vasomotor reactivity: steady-state versus transient changes in carbon dioxide tension. Exp Physiol 99, 1499–1510. Brown MM & Pickles H (1982). Effect of epoprostenol ( prostacyclin , PGI2 ) on cerebral blood flow in man. J Neurol Neurosurg Psychiatry 45, 1033–1036. Bruhn H, Fransson P & Frahm J (2001). Modulation of cerebral blood oxygenation by indomethacin: MRI at rest and functional brain activation. J Magn Reson Imaging 13, 325–334. Brune K & Patrignani P (2015). New insights into the use of currently available non-steroidal anti-inflammatory drugs. J Pain Res 8, 105–118. Burgess K, Lucas S, Sheperd K, Dawson A, Swart M, Thomas K, Lucas R, Donnelly P, Peebles K, Basnyat R & Ainslie P (2014). Influence of cerebral blood flow on central sleep apnea at high altitude. Sleep. ! 76!Burgess KR, Fan J, Peebles K, Thomas K, Lucas S, Lucas R, Dawson A, Swart M, Shepherd K & Ainsl (2010). Exacerbation of Obstructive Sleep Apnea by Oral Indomethacin. Chest 137, 707–710. Busija D (1983). Role of Prostaglandins in the Response of the Cerebral Circulation to Carbon Dioxide in Conscious Rabbits. J Cereb Blood Flow Metab376–380. Busija D & Heistad D (1983). Effects of indomethacin on cerebral blood flow during hypercapnia in cats. Am J Physiol Heart Circ Physiol 244, H519–H524. Busse R, Förstermann U, Matsuda H & Pohl U (1984). The role of prostaglandins in the endothelium-mediated vasodilatory response to hypoxia. Pflugers Arch 401, 77–83. Cassaglia P a, Griffiths RI & Walker AM (2008). Sympathetic nerve activity in the superior cervical ganglia increases in response to imposed increases in arterial pressure. Am J Physiol Regul Integr Comp Physiol 294, R1255–R1261. Chemtob S, Beharry K, Barna T, Varma DR & Aranda J V (1991). Differences in the effects in the newborn piglet of various nonsteroidal antiinflammatory drugs on cerebral blood flow but not on cerebrovascular prostaglandins. Pediatr Res 30, 106–111. Coverdale NS, Gati JS, Opalevych O, Perrotta A & Shoemaker JK (2014). Cerebral blood flow velocity underestimates cerebral blood flow during modest hypercapnia and hypocapnia. J Appl Physiol 117, 1090–1096. Coverdale NS, Lalande S, Perrotta A & Shoemaker JK (2015). Hetergeneous patterns of vasoreactivity in the middle cerebral and internal carotid arteries. Am J Physiol Heart Circ Physiolajpheart.00761.2014. Davis RJ, Murdoch CE, Ali M, Purbrick S, Ravid R, Baxter GS, Tilford N, Sheldrick RLG, Clark KL & Coleman RA (2004). EP4 prostanoid receptor-mediated vasodilatation of human middle cerebral arteries. Br J Pharmacol 141, 580–585. Dempsey JA, Veasey SC, Morgan BJ & O’Donnell CP (2010). Pathophysiology of sleep apnea. Physiol Rev47–112. Ellis EF, Police RJ, Yancey L, McKinney JS & Amruthesh SC (1990). Dilation of cerebral arterioles by cytochrome p-450 metabolites of arachadonic acid. Am J Physiol Heart Circ Physiol 259, 1171–1177. Ellis EF, Wei EP, Cockrell CS, Traweek DL, Saady JJ & Kontos HA (1980). The Effect of O2 and CO2 on Prostaglandin Levels in the Cat Cerebral Cortex. Circ Res 51, 652–657. Ellis EF, Wei EP & Kontos A (1979). Vasodilation of cat cerebral prostaglandins arterioles by prostaglandins D2, E2, G2, and I2. Am J Physiol Heart Circ Physiol 6, H381–H385. ! 77!Eriksson S, Hagenfeldt L, Law D, Patrono C, Pinca E & Wennmalm A (1983). Effect of prostaglandin synthesis inhibitors on basal and carbon dioxide stimulated cerebral blood flow in man. Acta Physiol Scand 117, 203–211. Evans D (1985). On the measurement of the mean velocity of blood flow over the cardiac cycle using Doppler ultrasound. Ultrasound Med Biol 11, 735–741. Fan JL, Burgess KR, Thomas KN, Peebles KC, Lucas SJE, Lucas RAI, Cotter JD & Ainslie PN (2010). Influence of indomethacin on ventilatory and cerebrovascular responsiveness to CO2 and breathing stability: the influence of PCO2 gradients. AJP Regul Integr Comp Physiol 298, R1648–R1658. Faraci FM & Heistad DD (1990). Regulation of Large Cerebral Arteries and Cerebral Microvascular Pressure. Circ Res 66, 8–17. Faraci FM & Sobey CG (1998). Role of potassium channels in regulation of cerebral vascular tone. J Cereb blood flow Metab 18, 1047–1063. Fierstra J, Sobczyk O, Battisti-Charbonney A, Mandell DM, Poublanc J, Crawley AP, Mikulis DJ, Duffin J & Fisher JA (2013). Measuring cerebrovascular reactivity: what stimulus to use? J Physiol 591, 5809–5821. Fog M (1937). Cerebral circulation: the reaction of the pial arteries to a fall in blood pressure. Arch Neurol Psychiatry 37, 351–364. Fog M (1939). Cerebral circulation: II. Reaction of pial arteries to increase in blood pressure. Arch Neurol Psychiatry 41, 260–268. Gecse A, Ottlecz A, Mezei Z, Telegdy G & Joo F (1982). Prostacyclin and prostaglandin synthesis in isolated brain capillaries. Prostaglandins. Giller C (2003). The Emperor Has No Clothes: Velocity, Flow, and the Use of TCD. J Neuroimaging 13, 97–98. Giller CA, Bowman G, Dyer H, Mootz L & Krippner W (1993). Cerebral arterial diameters during changes in blood pressure and carbon dioxide during craniotomy. Neurosurgery 32, 732–737. Gordon GRJ, Choi HB, Rungta RL, Ellis-Davies GCR & MacVicar BA (2008). Brain metabolism dictates the polarity of astrocyte control over arterioles. Nature 456, 745–749. Green DJ, Dawson EA, Groenewoud HMM, Jones H & Thijssen DHJ (2014). Is flow-mediated dilation nitric oxide mediated?: A meta-analysis. Hypertension 63, 376–382. ! 78!Green RS, Leffler CW, Busija DW, Fletcher AM & Beasley DG (1987). Indomethacin does not alter the circulating catecholamine response to asphyxia in the neonatal piglet. Pediatr Res 21, 534–537. Gupta A, Chazen JL, Hartman M, Delgado D, Anumula N, Shao H, Mazumdar M, Segal AZ, Kamel H, Leifer D & Sanelli PC (2012). Cerebrovascular reserve and stroke risk in patients with carotid stenosis or occlusion: A systematic review and meta-analysis. Stroke 43, 2884–2891. Hagen AA, White RP & Robertson JT (1979). Synthesis of prostaglandins and thromboxane B2 by cerebral arteries. Stroke 10, 306–309. Harper a M & Glass HI (1965). Effect of alterations in the arterial carbon dioxide tension on the blood flow through the cerebral cortex at normal and low arterial blood pressures. J Neurol Neurosurg Psychiatry 28, 449–452. Harper AM & Bell RA (1963). The effect of metabolic acidosis and alkalosis on the blood flow through the cerebral cortex. J Neurol Neurosurg Psychiatry 471, 341–344. Hashimoto H, Negishi M & Ichikawa A (1990). Identification of a prostacyclin receptor coupled to the adenylate cyclase system via a stimulatory GTP-binding protein in mouse mastocytoma P-815 cells. Prostaglandins 40, 491–505. Hauge A, Thoresen M & Wallge L (1980). Changes in cerebral blood flow during hyperventilation and C02-breathing measured transcutaneously in humans by a bidirectional, pulsed, ultrasound Doppler blood velocitymeter. Acta Physiol Scand 110, 167–173. Heistad DD, Marcus ML & Abboud FM (1978). Role of large arteries in regulation of cerebral blood flow in dogs. J Clin Invest 62, 761. Hirshman C, McCullough R & Weil J V (1975). Normal values for hypoxic and hypercapnic ventilatory drives in man. J Appl Physiol 38, 1095–1098. Hoiland RL, Ainslie PN, Wildfong KW, Smith KJ, Bain AR, Willie CK, Foster G, Monteleone B & Day TA (2015). Indomethacin-induced impairment of regional cerebrovascular reactivity: implications for respiratory control. J Physiol 593, 1291–1306. Hoiland RL, Day TA, Wildfong KW, Smith KJ, Bain AR, Willie CK, Foster GE, Monteleone B & Ainslie PN (2014). Hypercapnia induces dilation of large cerebral arteries and is mediated via a non-selective cyclooxygenase pathway (LB704). FASEB J. Hsu P, Shibata M & Leffler C (1993). Prostanoid synthesis in response to high CO2 in newborn pig brain microvascular endothelial cells. Am J … 264, H1485–H1492. ! 79!Iadecola C & Zhang F (1996). Permissive and obligatory responses of NO in cerebrovascular responses to hypercapnia and acetylcholine. Am J Physiol Regul Integr Comp Physiol 271, R990–R1001. Ide K, Boushel R, Sérensen HM, Fernandes A, Cai Y & Pott F (2000). Middle cerebral artery blood velocity during exercise with beta-1 adrenergic and unilateral stellate ganglion blockade in humans. Acta Physiol Scand 170, 33–38. Ide K, Eliasziw M & Poulin MJ (2003). Relationship between middle cerebral artery blood velocity and end-tidal PCO2 in the hypocapnic-hypercapnic range in humans. J Appl Physiol 95, 129–137. Ide K, Worthley M, Anderson T & Poulin MJ (2007). Effects of the nitric oxide synthase inhibitor L-NMMA on cerebrovascular and cardiovascular responses to hypoxia and hypercapnia in humans. J Physiol 584, 321–332. Inaba Y, Chen JA & Bergmann SR (2010). Prediction of future cardiovascular outcomes by flow-mediated vasodilatation of brachial artery: a meta-analysis. Int J Cardiovasc Imaging 26, 631–640. Jallad NS, Garg DC, Martinez JJ, Mroszczak EJ & Weidler DJ (1990). Pharmacokinetics of single-dose oral and intramuscular Ketorolac Tromethamine in the Young and elderly. J Clin Pharmacol 30, 76–81. Javaheri S & Dempsey JA (2013). Central sleep apnea. Compr Physiol 3, 141–163. Jordan J, Shannon JR, Diedrich a., Black B, Costa F, Robertson D & Biaggioni I (2000). Interaction of Carbon Dioxide and Sympathetic Nervous System Activity in the Regulation of Cerebral Perfusion in Humans. Hypertension 36, 383–388. Jung D, Mroszczak EJ, Wu A, Ling TL, Sevelius H & Bynum L (1989). Pharmacokinetics of Ketorolac and p-hydroxyketorolac following oral administration of ketorolac tromethamine. Pharm Res 6, 62–65. Kang CK, Oh ST, Chung RK, Lee H, Park CA, Kim YB, Yoo JH, Kim DY & Cho ZH (2010). Effect of stellate ganglion block on the cerebrovascular system: magnetic resonance angiography study. Anesthesiology 113, 936–944. Kantor H & Hampton M (1978). Indomethacin in submicromolar concentrations inhibits cyclic AMP-dependent protein kinase. Nature 276, 841–842. Kasischke KA, Vishwasrao HD, Fisher PJ, Zipfel WR & Webb WW (2004). Neural Activity Triggers Neuronal Oxidative Metabolism Followed by Astrocytic Glycolysis. Science (80- ) 305, 99–103. ! 80!Kastrup A, Happe V, Hartmann C & Schabet M (1999). Gender-related effects of indomethacin on cerebrovascular CO2 reactivity. J Neurol Sci 162, 127–132. Kerrick WG & Hoar PE (1981). Inhibition of smooth muscle tension by cyclic AMP-dependent protein kinase. Nature 292, 253–255. Kety SS & Schmidt CF (1948a). The nitrous oxide method for the quantitative determination of cerebral blood flow in man: theory, procedure and normal values. J Clin Invest 27, 476–483. Kety SS & Schmidt CF (1948b). The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 27, 484–492. Kontos H a., Wei EP, Ellis EF, Jenkins LW, Povlishock JT, Rowe GT & Hess ML (1985). Appearance of superoxide anion radical in cerebral extracellular space during increased prostaglandin synthesis in cats. Circ Res 57, 142–151. Kontos H a., Wei EP, Povlishock JT & Christman CW (1984). Oxygen radicals mediate the cerebral arteriolar dilation from arachidonate and bradykinin in cats. Circ Res 55, 295–303. Kontos H, Wei E, Navari RM, Levasseur JE, Rosenblum WI & Patterson JL (1978). Responses of cerebral arteries and arterioles to acute hypotension and hypertension. Am J Physiol Heart Circ Physiol 234, H371–H383. Kontos HA, Raper AJ & Patterson JL (1977a). Analysis of vasoactivity of local pH, PCO2 and bicarbonate on pial vessels. Stroke 8, 358–360. Kontos HA, Wei EP, Raper AJ & Patterson JL (1977b). Local mechanism of CO2 action of cat pial arterioles. Stroke 8, 226–229. Kowiański P, Lietzau G, Steliga A, Waśkow M & Moryś J (2013). The astrocytic contribution to neurovascular coupling--still more questions than answers? Neurosci Res 75, 171–183. Kraaier V, Huffelen AC Van, Wieneke GH, Van Der Worp HB & Bär PR (1992). Quantitative EEG changes due to cerebral vasoconstriction. Indomethacin versus hyperventilation-induced reduction in cerebral blood flow in normal subjects. Electroencephalogr Clin Neurophysiol 82, 208–212. Kurzrok R & Lieb CC (1930). Biochemical studies on human semen. II. The action of semen on the human uterus. Exp Biol Med 28, 268–272. Lambertsen CJ, Gelfand R & Semple SJG (1961). H+ and pC02, as chemical factors in respiratory and cerebral circulatory control. J Appl Physiol 16, 473–484. ! 81!Lassen NA (1959). Cerebral blood flow and oxygen consumption in man. Physiol Rev 39, 183–238. Leffler C & Busija D (1985). Prostanoids in cortical subarachnoid cerebrospinal fluid and pial arterial diameter in newborn pigs. Circ Res 57, 689–694. Leffler C & Busija D (1987). Prostanoids and pial arteriolar diameter in hypotensive newborn pigs. Am J Physiol Heart Circ Physiol 252, 687–691. Leffler C, Mirro R, Pharris LJ & Shibata M (1994). Permissive role of prostacyclin in cerebral vasodialtion to hypercapnia in newborn pigs. Am J Physiol Heart Circ Physiol 267, H285–H291. Leffler CW, Balabanova L & Williams KK (1999). cAMP production by piglet cerebral vascular smooth muscle cells): pHo, pHi, and permissive action of PGI2. Am J Physiol Heart Circ Physiol 277, H1878–H1883. Leffler CW, Busija DW, Armstead WM, Shanklin DR, Mirro R & Thelin O (1990). Activated oxygen and arachidonate on newborn cerebral arterioles. Am J Physiol Heart Circ Physiol 259, 1230–1238. Leffler CW & Fedinec AL (1997). Newborn piglet cerebral microvascular responses to epoxyeicosatrienoic acids. Am J Physiol Heart Circ Physiol 273, H333–H338. Leffler CW, Mirro R, Shibata M, Parfenova H, Armstead WM & Zuckerman S (1993). Effects of indomethacin on cerebral vasodilator responses to arachidonic acid and hypercapnia in newborn pigs. Pediatr Res 33, 609–614. Leffler CW, Mirro R, Thompson C, Shibata M, Armstead WM, Pourcyrous M & Thelin O (1991). Activated oxygen species do not mediate hypercapnia-induced cerebral vasodilation in newborn pigs. Am J Physiol Circ Physiol 261, H335–H342. LeMarbre G, Stauber S, Khayat RN, Puleo DS, Skatrud JB & Morgan BJ (2003). Baroreflex-induced sympathetic activation does not alter cerebrovascular CO2 responsiveness in humans. J Physiol 551, 609–616. Lennox WG & Gibbs EL (1932). The blood flow in the brain and the leg of man, and the changes induced by alteration of blood gases. J Clin Invest1155–1177. Lennox WG, Gibbs FA & Gibbs EL (1935). Relationship of unconsciousness to cerebral blood flow and to anoxemia. Arch Neurol Psychiatry 34, 1001–1013. Lewis N, Smith K, Bain AR, Wildfong KW, Numan T & Ainslie PN (2015). Impact of transient hypotension on regional cerebral blood flow in humans. Clin Sci. ! 82!Lewis NCS, Bain AR, MacLeod DB, Wildfong KW, Smith KJ, Willie CK, Sanders ML, Numan T, Morrison S a., Foster GE, Stewart JM & Ainslie PN (2014a). Impact of hypocapnia and cerebral perfusion on orthostatic tolerance. J Physiol 592, 5203–5219. Lewis NCS, Messinger L, Monteleone B & Ainslie PN (2014b). Effect of acute hypoxia on regional cerebral blood flow: effect of sympathetic nerve activity. J Appl Physiol 116, 1189–1196. Van Lieshout JJ, Wieling W, Karemaker JM & Secher NH (2003). Syncope, cerebral perfusion, and oxygenation. J Appl Physiol 94, 833–848. Lucas SJE, Tzeng YC, Galvin SD, Thomas KN, Ogoh S & Ainslie PN (2010). Influence of Changes in Blood Pressure on Cerebral Perfusion and Oxygenation. Hypertension 55, 698–705. Mardimae A, Balaban DY, Machina M a, Battisti-Charbonney A, Han JS, Katznelson R, Minkovich LL, Fedorko L, Murphy PM, Wasowicz M, Naughton F, Meineri M, Fisher J a & Duffin J (2012). The interaction of carbon dioxide and hypoxia in the control of cerebral blood flow. Pflugers Arch 464, 345–351. Markus H & Cullinane M (2001). Severely impaired cerebrovascular reactivity predicts stroke and TIA risk in patients with carotid artery stenosis and occlusion. Brain457–467. Markus HS, Vallance P & Brown MM (1994). Differential effect of three cyclooxygenase inhibitors on human cerebral blood flow velocity and carbon dioxide reactivity. Stroke 25, 1760–1764. Marsh EE, Shaw ND, Klingman KM, Tiamfook-Morgan TO, Yialamas M a., Sluss PM & Hall JE (2011). Estrogen levels are higher across the menstrual cycle in African-American women compared with Caucasian women. J Clin Endocrinol Metab 96, 3199–3206. Meng L & Gelb AW (2015). Regulation of cerebral autoregulation by carbon dioxide. Anesthesiology 122, 196–205. Messina EJ, Sun D, Koller A, Wolin MS & Kaley G (1992). Role of endothelium-derived prostaglandins in hypoxia-elicited arteriolar dilation in rat skeletal muscle. Circ Res 71, 790–796. Miyazaki M & Kato K (1965). Measurement of cerebral blood flow by ultrasonic doppler technique; hemodynamic comparison of right and left carotid artery in patients with hemiplegia. Jpn Circ J 29, 383–386. ! 83!Moncada S, Gryglewski RJ, Bunting S & Vane JR (1976). An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Mosso A (1880). Sulla circolazione del cervello dell’uomo. Att R Accad Lincei 5, 237–358. Narumiya S, Sugimoto Y & Ushikubi F (1999). Prostanoid Receptors: Structures, Properties, and Functions. Physiol Rev 79, 1193–1227. Nicholson WT, Vaa B, Hesse C, Eisenach JH & Joyner MJ (2009). Aging Is Associated With Reduced Prostacyclin-Mediated Dilation in the Human Forearm. Hypertension 53, 973–978. Niwa K, Haensel C, Ross ME & Iadecola C (2001). Cyclooxygenase-1 Participates in Selected Vasodilator Responses of the Cerebral Circulation. Circ Res 88, 600–608. Nowak J & Wennmalm a (1978). Influence of indomethacin and of prostaglandin E1 on total and regional blood flow in man. Acta Physiol Scand 102, 484–491. Numan T, Bain AR, Hoiland RL, Smirl JD, Lewis NC & Ainslie PN (2014). Static autoregulation in humans: a review and reanalysis. Med Eng Phys 36, 1487–1495. Offenhauser N, Thomsen K, Caesar K & Lauritzen M (2005). Activity-induced tissue oxygenation changes in rat cerebellar cortex: interplay of postsynaptic activation and blood flow. J Physiol 565, 279–294. Ogawa S, Handa N, Matsumoto M, Etani H, Yoneda S, Kimura K & Kamada T (1988). Carbon dioxide reactivity of the blood flow in human basilar artery estimated by the transcranial doppler method in normal men: a comparison with that of the middle cerebral artery. Ultrasound Med Biol 14, 479–483. Ospina J a, Duckles SP & Krause DN (2003). 17beta-estradiol decreases vascular tone in cerebral arteries by shifting COX-dependent vasoconstriction to vasodilation. Am J Physiol Heart Circ Physiol 285, H241–H250. Pandit JJ, Mohan RM, Paterson ND & Poulin MJ (2003). Cerebral blood flow sensitivity to CO2 measured with steady-state and Read’s rebreathing methods. Respir Physiol Neurobiol 137, 1–10. Pandit JJ, Mohan RM, Paterson ND & Poulin MJ (2007). Cerebral blood flow sensitivities to CO2 measured with steady-state and modified rebreathing methods. Respir Physiol Neurobiol 159, 34–44. Parfenova H, Hsu P & Leffler CW (1995a). Dilator Prostanoid-induced cyclic AMP formation and release by Cerebral Microvascular Smooth Muscle Cells: Inhibition by indomethacin. J Pharmacol Exp Ther 272, 44–52. ! 84!Parfenova H & Leffler C (1996). Effects of hypercapnia on prostanoid and cAMP production by cerebral microvascular cell cultures. Am J Physiol Cell Physiol 270, C1503–C1510. Parfenova H, Shibata M, Zuckerman S & Leffler CW (1994). CO2 and cerebral circulation in newborn pigs: cyclic nucleotides and prostanoids in vascular regulation. Am J Physiol Heart Circ Physiol 266, H1494–H1501. Parfenova H, Shibata M, Zuckerman S, Mirro R & Leffler CW (1993). cyclic nucleotides and cerebrovascular tone in newborn pigs. Am J Physiol Heart Circ Physiol 265, H1972–H1982. Parfenova H, Zuckerman S & Leffler CW (1995b). Inhibitory effect of indomethacin on prostacyclin receptor-mediated cerebral vascular responses. Am J Physiol Heart Circ Physiol 268, H1884–H1890. Parsons AA & Whalley ET (1989). Effects of Prostanoids on Human and Rabbit Basilar Arteries Precontracted In Vitro. Cephalalgia 9, 165–171. Paul KS, T WE, Forster C, Lye R & Dutton J (1982). Prostacyclin and cerebral vessel relaxation. J Neurosurg 57, 334–340. Peebles K, Celi L, McGrattan K, Murrell C, Thomas K & Ainslie PN (2007). Human cerebrovascular and ventilatory CO2 reactivity to end-tidal, arterial and internal jugular vein PCO2. J Physiol 584, 347–357. Peebles KC, Ball OG, MacRae B a, Horsman HM & Tzeng YC (2012). Sympathetic regulation of the human cerebrovascular response to carbon dioxide. J Appl Physiol 113, 700–706. Perea G, Navarrete M & Araque A (2009). Tripartite synapses: astrocytes process and control synaptic information. Trends Neurosci 32, 421–431. Phillis JW & DeLong RE (1987). An involvement of adenosine in cerebral blood flow regulation during hypercapnia. Gen Pharmacol 18, 133–139. Pickard J & MacKenzie E (1973). Inhibition of prostaglandin synthesis and the response of baboon cerebral circulation to carbon dioxide. Nature 245, 187–188. Pickard J, Tamura A, Stewart M, McGeorge A & Fitch W (1980). Prostacycline, indomethacin and the cerebral circulation. Brain Res 197, 425–431. Pickles H, Brown MM, Thomas M, Hewazy a H, Redmond S, Zilkha E & Marshall J (1984). Effect of indomethacin on cerebral blood flow, carbon dioxide reactivity and the response to epoprostenol (prostacyclin) infusion in man. J Neurol Neurosurg Psychiatry 47, 51–55. ! 85!Portegies MLP, de Bruijn RFAG, Hofman A, Koudstaal PJ & Ikram MA (2014). Cerebral vasomotor reactivity and risk of mortality: the Rotterdam Study. Stroke 45, 42–47. Przybyłowski T, Bangash M-F, Reichmuth K, Morgan BJ, Skatrud JB & Dempsey J a (2003). Mechanisms of the cerebrovascular response to apnoea in humans. J Physiol 548, 323–332. Quan H & Shih WJ (1996). Assessing reproducibility by the within-subject coefficient of variation with random effects models. Biometrics 52, 1195–1203. Ramsay SC, Murphy K, Shea SA, Friston KJ, Lammertsma AA, Clark JC, Adams L, Guz A & Frackowiak RS (1993). Changes in global cerebal blood flow in humans: effect on regional cerebral blood flow during a neural activation task. J Physiol 471, 521–534. Regan RE, Fisher JA & Duffin J (2014). Factors affecting the determination of cerebrovascular reactivity. Brain Behav 4, 775–788. Rosenblum W (1983). Effects of free radical generation on mouse pial arterioles: probable role of hydroxyl radicals. Am J Physiol. Rota Kops E, Herzog H, Schmid A, Holte S & Feinendegen LE (1990). Performance of an eight-ring whole body PET scanner. J Comput Assist Tomogr 14, 437–445. Roy C & Sherrington C (1890). ON THE REGULATION OF THE BLOOD-SUPPLY OF the brain. J Physiol 11, 85–158. Samuelsson B (2012). Role of Basic Science in the Development of New Medicines: Examples from the Eicosanoid Field. J Biol Chem 287, 10070–10080. Sato K, Sadamoto T, Hirasawa A, Oue A, Subudhi AW, Miyazawa T & Ogoh S (2012). Differential blood flow responses to CO2 in human internal and external carotid and vertebral arteries. J Physiol 590, 3277–3290. Schmetterer L, Findl O, Strenn K, Graselli U, Kastner J, Eichler HG & Wolzt M (1997). Role of NO in the O2 and CO2 responsiveness of cerebral and ocular circulation in humans. Am J Physiol 273, R2005–R2012. Schöning M, Walter J & Scheel P (1994). Estimation of cerebral blood flow through color duplex sonography of the carotid and vertebral arteries in healthy adults. Stroke 25, 17–22. Seidel E, Eicke BM, Tettenborn B & Krummenauer F (1999). Reference values for vertebral artery flow volume by duplex sonography in young and elderly adults. Stroke 30, 2692–2696. ! 86!Serrador JM, Picot PA, Rutt BK, Shoemaker JK & Bondar RL (2000). MRI Measures of Middle Cerebral Artery Diameter in Conscious Humans During Simulated Orthostasis. Stroke 31, 1672–1678. Shapiro W, Wasserman AJ & Patterson JL (1966). Mechanism and Pattern of Human Cerebrovascular Regulation after Rapid Changes in Blood CO2 Tension. J Clin Invest. Skow RJ, MacKay CM, Tymko MM, Willie CK, Smith KJ, Ainslie PN & Day TA (2013). Differential cerebrovascular CO2 reactivity in anterior and posterior cerebral circulations. Respir Physiol Neurobiol 189, 76–86. Smith B a, Clayton EW & Robertson D (2011). Experimental arrest of cerebral blood flow in human subjects: the red wing studies revisited. Perspect Biol Med 54, 121–131. Smith JJ, Hillard CJ, Lee JG, Hudetz AG, Bosnjak ZJ & Kampine JP (1997). The Role of Nitric Oxide in the Cerebrovascular to Hypercapnia. Anesth Analg 84, 363–369. Smith KJ, Wong LE, Eves ND, Koelwyn GJ, Smirl JD, Willie CK & Ainslie PN (2012). Regional cerebral blood flow distribution during exercise: influence of oxygen. Respir Physiol Neurobiol 184, 97–105. Smith WL (1992). Prostanoid biosynthesis and mechanisms of action. Am J Physiol 263, F181–F191. Smith WL, Marnett LJ & DeWitt DL (1991). prostaglandin and thromboxane biosynthesis. Pharmacol Ther. Song Y & Simard M (1995). Beta-Adrenoceptor stimulation activates large. Conductance Ca2+ -activated K+ channels in smooth muscle cells from basilar artery of guinea pig. Pflugers Arch 430, 984–993. St Lawrence KS, Ye FQ, Lewis BK, Frank JA & Mclaughlin AC (2003). Measuring the effects of indomethacin on changes in cerebral oxidative metabolism and cerebral blood flow during sensorimotor activation. Magn Reson Med 50, 99–106. St Lawrence KS, Ye FQ, Lewis BK, Weinberger DR, Frank JA & McLaughlin AC (2002). Effects of indomethacin on cerebral blood flow at rest and during hypercapnia: an arterial spin tagging study in humans. J Magn Reson Imaging 15, 628–635. Staessen J, Cattaert AN, Fagard R, Lijnen P, Moerman E, De Schaepdryver A & Amery A (1984). Hemodynamic and humoral effects of prostaglandin inhibition in exercising humans. J Appl Physiol 56, 39–45. Statland BE & Demas TJ (1980). Serum caffeine half-lives. Healthy subjects vs. patients having alcoholic hepatic disease. Am J Clin Pathol 73, 390–393. ! 87!Steinback CD, Breskovic T, Banic I, Dujic Z & Shoemaker JK (2010a). Autonomic and cardiovascular responses to chemoreflex stress in apnoea divers. Auton Neurosci 156, 138–143. Steinback CD, Breskovic T, Frances M, Dujic Z & Shoemaker JK (2010b). Ventilatory restraint of sympathetic activity during chemoreflex stress. 1407–1414. Steinback CD, Salzer D, Medeiros PJ, Kowalchuk J & Shoemaker JK (2009). Hypercapnic vs. hypoxic control of cardiovascular, cardiovagal, and sympathetic function. Am J Physiol Regul Integr Comp Physiol 296, R402–R410. Szabo K, Rosengarten B, Juhasz T, Lako E, Csiba L & Olah L (2014). Effect of non-steroid anti-inflammatory drugs on neurovascular coupling in humans. J Neurol Sci 336, 227–231. Tan CO (2012). Defining the characteristic relationship between arterial pressure and cerebral flow. J Appl Physiol 113, 1194–1200. Tian L (2006). Inferences on the within-subject coefficient of variation. Stat Med 25, 2008–2017. Toda N & Miyazaki M (1978). Responses of isolated dog cerebral and peripheral arteries to prostaglandins after application of aspirin and polyphloretin phosphate. Stroke 9, 490–498. Tymko MM, Ainslie PN, Macleod DB, Willie CK & Foster GE (2015). End-tidal-to-arterial CO2 and O2 gas gradients at low- and high-altitude during dynamic end-tidal forcing. Am J Physiol - Regul Integr Comp Physiol; DOI: 10.1152/ajpregu.00425.2014. Tzeng Y-C & Ainslie PN (2014). Blood pressure regulation IX: cerebral autoregulation under blood pressure challenges. Eur J Appl Physiol 114, 545–559. Tzeng Y-C, Willie CK, Atkinson G, Lucas SJE, Wong A & Ainslie PN (2010). Cerebrovascular regulation during transient hypotension and hypertension in humans. Hypertension 56, 268–273. Umeyama T, Kugimiya T, Ogawa T, Kandori Y, Ishizuka A & Hanaoka K (1995). Changes in cerebral blood flow estimated after stellate ganglion block by single photon emission computed tomography. J Auton Nerv Syst 50, 339–346. Uski T, Andersson K, Brandt L, Edvinsson L & Ljunggren B (1983). Responses of Isolated Feline and Human Cerebral Arteries to Prostacycliri and Some of Its Metabolites. 238–245. Valdueza JM, Balzer JO, Villringer A, Vogl TJ, Kutter R & Einhäupl KM (1997). Changes in blood flow velocity and diameter of the middle cerebral artery during ! 88!hyperventilation: assessment with MR and transcranial Doppler sonography. Am J Neuroradiol 18, 1929–1934. Vanzetta I & Grinvald A (1999). Increased Cortical Oxidative Metabolism Due to Sensory Stimulation: Implications for Functional Brain Imaging. Science (80- ) 286, 1555–1559. Verbree J, Bronzwaer AGT, Ghariq E, Versluis MJ, Daeman MJAP, van Buchem MA, Dahan A, van Lieshout JJ & van Osch MJP (2014). Assessment of middle cerebral artery diameter during hypocapnia and hypercapnia in humans using ultra high-field MRI. J Appl Physiol; DOI: 10.1152/japplphysiol.00651.2014. Wagerle LC & Mishra OP (1988). Mechanism of CO2 response in cerebral arteries of the newborn pig: role of phospholipase, cyclooxygenase, and lipoxygenase pathways. Circ Res 62, 1019–1026. Wahl M, Kuschinsky W, Bosse O & Thurau K (1973). Dependency of Pial Arterial and Arteriolar on Perivascular Osmolarity in the Cat: A microapplication study. Circ Res 32, 162–170. Wahl M, Schilling L & Whalley ET (1989). Cerebrovascular effects of prostanoids: In-situ studies in pial arteries of the cat. Naunyn Schmiedebergs Arch Pharmacol 340, 314–320. Wang Q, Paulson OB & Lassen NA (1993). Indomethacin Abolishes Cerebral Blood Flow Increase in Response to Acetazolamide-Induced Extracellular Acidosis): A Mechanism for Its Effect on Hypercapnia? J Cereb Blood Flow Metab 13, 724–727. Wang Q, Pelligrino DA, Koenig HM & Albrecht RF (1994a). The role of endothelium and nitric oxide in rat pial arteriolar dilatory responses to CO2 in vivo. J Cereb blood flow Metab 14, 944–951. Wang Q, Pelligrino DA, Paulson OB & Lassen NA (1994b). Comparison of the effects of NG-nitro-L-arginine and indomethacin on the hypercapnic cerebral blood flow increase in rats. Brain Res 641, 257–264. Wang X, Wu J, Li L, Chen F, Wang R & Jiang C (2003). Hypercapnic acidosis activates KATP channels in vascular smooth muscles. Circ Res 92, 1225–1232. Wei E, Ellis E & Kontos H (1980). Role of prostaglandins in pial arteriolar response to CO2 and hypoxia. Am J Physiol Heart Circ Physiol 238, H226–H230. Wei EP, Christman CW, Kontos HA & Povlishock JT (1985). Effects of oxygen radicals on cerebral arterioles. Am J Physiol 248, 157–162. Weksler BB, Marcus AJ & Jaffe EA (1977). Synthesis of prostaglandin I2 (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Acad Sci 74, 3922–3926. ! 89!Welch K, Knowles L & Spira P (1974). local effect of prostaglandins on cat pial arteries. Eur J Pharmacol 25, 155–158. Wennmalm A, Carlsson I, Edlund A, Eriksson S, Kaijser L & Nowak J (1984). Central and peripheral haemodynamic effects of non-steroidal anti-inflammatory drugs in man. Arch Toxicol Suppl 7, 350–359. Wennmalm Å, Eriksson S & Wahren J (1981). Effect of indomethacin on basal and carbon dioxide stimulated cerebral blood flow in man. Clin Physiol 1, 227–234. Whalley ET, Schilling L & Wahl M (1989). Cerebrovascular Effects of Prostanoids: IN-VITRO Studies in Middle Cerebral and Basilar Artery. Prostaglandins 38, 625–634. White RP, Deane C, Vallance P & Markus HS (1998). Nitric Oxide Synthase Inhibition in Humans Reduces Cerebral Blood Flow but Not the Hyperemic Response to Hypercapnia. Stroke 29, 467–472. Willie CK, Colino FL, Bailey DM, Tzeng YC, Binsted G, Jones LW, Haykowsky MJ, Bellapart J, Ogoh S, Smith KJ, Smirl JD, Day TA, Lucas SJ, Eller LK & Ainslie PN (2011a). Utility of transcranial Doppler ultrasound for the integrative assessment of cerebrovascular function. J Neurosci Methods 196, 221–237. Willie CK, Cowan EC, Ainslie PN, Taylor CE, Smith KJ, Sin PYW & Tzeng YC (2011b). Neurovascular coupling and distribution of cerebral blood flow during exercise. J Neurosci Methods 198, 270–273. Willie CK, Macleod DB, Shaw AD, Smith KJ, Tzeng YC, Eves ND, Ikeda K, Graham J, Lewis NC, Day TA & Ainslie PN (2012). Regional brain blood flow in man during acute changes in arterial blood gases. J Physiol 590, 3261–3275. Willie CK, Tzeng Y-C, Fisher JA & Ainslie PN (2014). Integrative regulation of human brain blood flow. J Physiol 592, 841–859. Willis T (1664). Cerebri Anatome: Cui accessit nervorum description et usus (thesis). Wilson MH, Edsell MEG, Davagnanam I, Hirani SP, Martin DS, Levett DZH, Thornton JS, Golay X, Strycharczuk L, Newman SP, Montgomery HE, Grocott MPW & Imray CHE (2011). Cerebral artery dilatation maintains cerebral oxygenation at extreme altitude and in acute hypoxia - an ultrasound and MRI study. J Cereb Blood Flow Metab 31, 2019–2029. Wolff HG & Lennox WG (1930). Cerebral circulation: XII. The effect on pial vessels of variations in the oxygen and carbon dioxide content of the blood. Arch Neurol Psychiatry 23, 1097. ! 90!Woodman RJ, Playford DA, Watts GF, Cheetham C, Reed C, Taylor RR, Puddey IB, Beilin LJ, Burke V, Mori TA & Green DJ (2001). Improved analysis of brachial artery ultrasound using a novel edge-detection software system. J Appl Physiol929–937. Xie A, Skatrud JB, Barczi SR, Reichmuth K, Morgan BJ, Mont S & Dempsey J a (2009). Influence of cerebral blood flow on breathing stability. J Appl Physiol 106, 850–856. Xie A, Skatrud JB, Khayat R, Dempsey JA, Morgan B & Russell D (2005). Cerebrovascular response to carbon dioxide in patients with congestive heart failure. Am J Respir Crit Care Med 172, 371–378. Xie A, Skatrud JB, Morgan B, Chenuel B, Khayat R, Reichmuth K, Lin J & Dempsey JA (2006). Influence of cerebrovascular function on the hypercapnic ventilatory response in healthy humans. J Physiol 577, 319–329. Zarrinkoob L, Ambarki K, Wåhlin A, Birgander R, Eklund A & Malm J (2015). Blood flow distribution in cerebral arteries. J Cereb blood flow Metab1–7. Zhang R & Levine BD (2007). Autonomic ganglionic blockade does not prevent reduction in cerebral blood flow velocity during orthostasis in humans. Stroke 38, 1238–1244. Zwanenburg JJM, Versluis MJ, Luijten PR & Petridou N (2011). Fast high resolution whole brain T2* weighted imaging using echo planar imaging at 7T. Neuroimage 56, 1902–1907.  !

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0166400/manifest

Comment

Related Items