Secondary Logo

Journal Logo

Anaesthesia

Effects on cerebral blood flow of position changes, hyperoxia, CO2 partial pressure variations and the Valsalva manoeuvre

A study in healthy volunteers

Tercero, Javier; Gracia, Isabel; Hurtado, Paola; de Riva, Nicolás; Carrero, Enrique; Garcia-Orellana, Marta; Belda, Isabel; Rios, José; Maldonado, Felipe; Fàbregas, Neus; Valero, Ricard

Editor(s): 2020, Published online 16 October

Author Information
doi: 10.1097/EJA.0000000000001356

Abstract

Introduction

Maintaining adequate blood pressure (BP) to ensure proper brain perfusion during surgery is challenging.1 Some intra-operative circumstances are unavoidable (e.g. haemorrhage), others depend on surgical needs (e.g. induced hypotension in neuroendoscopic skull-base surgery), and others may be conditioned by positional changes (sitting), or intra-operative manoeuvres (Valsalva). Added to this, there are other factors which modulate cerebral blood flow (CBF), such as variations in carbon dioxide and oxygen tensions. However, the anaesthesiologist must maintain a haemodynamic situation that guarantees adequate CBF in any condition, but when several of these circumstances coincide, it is unclear how CBF is affected.2,3

CBF is regulated mainly through arteriolar dilation or contraction under the influence of multiple, complex physiological systems. Flow-metabolism coupling, neuronal activation, sympathetic nervous system activity and positioning also play a role.4–6 Moreover, changes in PaCO2 alter the relative volumes of arterial and venous blood in the brain; a 1.3-kPa (10-mmHg) decrease in PaCO2 can decrease the intracranial arterial blood volume by 50%.7

A head-up tilt position and hyperventilation are therapeutic strategies used to lower intracranial pressure. However, they carry a risk of inducing cerebral ischaemia if cerebral perfusion pressure and CBF decrease8–10 under the limits of autoregulation. Orthopaedic and neurosurgical procedures may require the sitting position. Severe cerebral ischaemia has been described in patients who underwent surgery in the beach-chair position even though adequate mean arterial pressure (MAP) had been maintained during the procedure.11

Noninvasive measures of regional cerebral oxygen saturation (rSO2) obtained by near-infrared spectroscopy (NIRS) seem to reflect a state of balance in oxygen delivery and uptake.12,13 If all parameters remain constant, a reduction in rSO2 due to a change in positioning will indicate a decrease in CBF because the systemic MAP has fallen below the level required for cerebral autoregulation. When Tisdall et al.14 used NIRS in healthy volunteers, decreases in the tissue oxygenation index after hypoxaemia (7.1%) and hypocapnia (2.1%) were observed, as well as increases after hyperoxia (2.3%) and hypercapnia (2.6%). The accuracy of cerebral oximeters has not been studied with varying position.15

We aimed primarily to evaluate CBF by means of its surrogates, transcranial Doppler ultrasound and NIRS, in healthy volunteers’ responses to standardised cardiac preload challenges; gravitational effects in supine, seated and standing positions; and during the Valsalva manoeuvre under conditions of both normal and altered ventilation.

Secondarily, we assessed whether this monitoring approach may help to detect alterations in cerebral autoregulation during some surgical procedures requiring the sitting position, leading to better patient-focused treatment.

Methods

The current study was approved by the research Ethics Committee of Hospital Clínic de Barcelona (Ref. 5000/2009, 23 April 2009). All participants responded to an IRB-approved advertisement and gave informed written consent. This article adheres to the applicable EQUATOR guidelines.

We enrolled 10 healthy nonsmoker volunteers, who did not drink beverages with stimulants (coffee, tea or soft drinks with caffeine) in the previous 12 h, and were allowed to take only a light meal 2 h before the experiment. The study took place in an operating room at a temperature of 22 °C. Monitoring included electrocardiography, heart rate (HR), peripheral blood oxygen saturation by pulse oximetry (probe on the right index finger) and noninvasive BP (S/5 Monitor; Datex-Ohmeda, Helsinki, Finland). With a sampling tube connected to nasal prongs and a Primus workstation (Dräger Medical, Lübeck, Germany), we monitored the following respiratory variables: end-tidal CO2 (ETCO2), fractions of inspired and expired oxygen (FiO2 and FeO2), and the fraction of inspired CO2. We also monitored transcutaneous partial pressure of CO2 (TcPCO2) in the left ear lobe (POC; Sentec, Therwil, Switzerland) and maximum and mean blood flow velocities (Vmax and Vmean, respectively) by transcranial Doppler ultrasound (Intraview; Rimed, Singen, Germany) in the right or left middle cerebral artery (MCA),16 depending on the quality of the insonation, at a constant insonation angle using a standard head-frame.

We measured the cardiac index (CI), total peripheral resistance index (TPRI) and stroke volume index (SVI) with a noninvasive cardiac output (CO) monitoring system (NICOM) (Reliant; Cheetah, Portland, Oregon, USA). The BP cuff was placed on the left arm. Right and left cerebral rSO2 and peripheral rSO2 at the right deltoid and quadriceps muscles were measured with the INVOS Oximeter 5100C (Somanetics, Troy, Michigan, USA). This monitor uses NIRS to estimate the balance between oxygen supply and demand. Two cold infrared light sources (730 and 810 nm) measured the proportion of oxyhaemoglobin and total haemoglobin concentrations and rSO2 is displayed as a percentage. To estimate cerebral rSO2, the light sensors were placed on the forehead with their proximal edges on the medial line and their caudal edges approximately 1 cm above the eyebrows; below the probe at this point the blood composition is approximately 20% arterial, 5% capillary and 75% venous.17 Baseline rSO2 values vary from 51 to 75% between individuals because of differences of anatomy and arterial, venous and capillary blood composition.7,17 For this study, we recorded the volunteers’ initial readings and thereafter the divergence from that reading. Since percentage changes are routinely recorded in clinical neuroanaesthesiological applications of the INVOS monitor, we recorded change as both the absolute difference between pretest and post test values and the percentage change.

The study included two sets of measurements. Figure 1 shows the timing and order of all the tests. The first set was designed to confirm increases in the CI occurring with cardiac preload-modifying position changes under physiological conditions. Specifically, we recorded the response to an ‘orthostatic challenge manoeuvre’ with the NICOM monitor's protocol (shift from seated to supine position) and to a ‘passive leg raise test’ with the volunteer seated at a 30° angle. The purpose of this first set under conditions of normal ventilation, with two manoeuvres to raise the CI, was to assess the integrity of the autonomic nervous system's fluid responsiveness in the volunteers. CI or SVI variations of more than 15% after increased cardiac preloading were considered positive results (denoting responsiveness). The NICOM system's ability to measure such changes noninvasively has been validated.18 We compared the NICOM, INVOS, and transcranial Doppler ultrasound results before and after the two preload-modifying manoeuvres (Fig. 1, top frame).

Fig. 1
Fig. 1:
Timing of data recording under physiological and experimental conditions. The orthostatic challenge protocol consisted of lying in the supine decubitus position after having been seated. The three positions in which responses were measured were supine decubitus, seated at a 50° angle and standing. The baseline measurements for each position were taken at the beginning of each set of experimental challenges. EEP, end-expiratory pressure; ETCO2, end-tidal carbon dioxide; F eO2, expiratory oxygen fraction; hC, hypocapnia; HC, hypercapnia; HO, hyperoxia; OC, orthostatic challenge; PLR, passive leg rise; V, Valsalva.

The second set of measurements recorded initially related to biological responses to gravitational changes in the supine, seated (as for posterior fossa surgery, trunk at a 50° angle) and standing positions (pretest measurements). These were followed by recordings in these positions during an experimental model of cerebral vasoconstriction (created by generating hyperoxia or hypocapnia) and a model of cerebral vasodilatation (by generating hypercapnia) (Fig. 1, lower frame). The experimental ventilatory conditions were as follows:

Hyperoxia (for vasoconstriction): after 100% oxygen was administered through a face mask for 3 min, the experimental condition was considered established if FeO2 values exceeded 70%. Controlled normal ventilation was then ensured by monitoring ETCO2 values to ensure there were no decreases of more than 0.67 kPa (5 mmHg) below baseline.

Hypocapnia (for vasoconstriction): the volunteers were asked to take rapid, deep breaths for as long as possible. The condition was considered established if ETCO2 decreased at least 1.3 kPa (10 mmHg) below baseline and held steady for 1 min.

Hypercapnia (for vasodilatation): the volunteers breathed, with a nose clip, in a hypercapnic (8 to 9% CO2) but not hypoxaemic (FiO2 0.3 to 0.35) gas mixture through a mouthpiece connected to a one-way T-valve. Inspired and expired gases were monitored while the volunteer breathed normally for as long as possible. The test condition was considered established if ETCO2 increased by at least 1.3 kPa over baseline and held steady for at least 1 min.

Finally, we recorded responses during a Valsalva manoeuvre performed in each position at the end of data recording in each of the three positions, before the volunteer changed position for the next group of measurements. We recorded variables at the same point in the manoeuvre for all volunteers (Fig. 1, lower frame). Uniform performance of the manoeuvre was ensured by asking the volunteers to inhale to vital capacity and then exhale as forcefully as possible through a mouthpiece connected to a manometer, sustaining an end-expiratory pressure (EEP) of more than 6.7 kPa (50 mmHg) with the nostrils occluded. The volunteer could see the pressure reading on the manometer at all times. The test was considered valid if the volunteer complied with those conditions and maintained the targeted EEP for at least 20 s.

Given the short, unpredictable duration of this manoeuvre, we recorded values just before the volunteer exhaled at the end of the increase of intrathoracic pressure (phase IIb of the Valsalva manoeuvre).19 HR begins to rise in phase IIa to compensate for a decrease in MAP. Just before exhalation, the HR is still rising but the increase in MAP can be detected in phase IIb. Given that the Valsalva manoeuvre precludes the recording of ETCO2, we also recorded TcPCO2 in all tests and this variable was used in all statistical analyses. We used ETCO2 to check the validity of test conditions because it is the variable customarily used to verify normocapnia, hypercapnia and hypocapnia in the literature. NICOM and INVOS readings and transcranial Doppler ultrasound results before and after each position and/or experimental condition and during the Valsalva manoeuvre were compared.

The experimental manoeuvres (hyperoxia, hypocapnia, hypercapnia and Valsalva) were taken consecutively during the three different positions (supine, seated, standing) in the same order for every participant.

Statistical analysis

Descriptive statistics are expressed as mean ± SD and mean percentage change. Possible influences of the different experimental situations were analysed using a generalised estimating equation technique with an autoregressive order 1 matrix to account for intra-individual correlations and differences. Period (pre, post) and baseline values were used as covariates to reflect changes due to positioning or the Valsalva manoeuvre. All analyses were performed using IBM SPSS Statistics for Windows (v. 20, IBM Corp, Armonk, New York, USA). A P value of less than 0.05 was considered significant for the analysis.

Results

Ten right-handed volunteers (five men and five women) aged between 25 and 38 years were enrolled between April and June 2010. Their mean weight was 68.9 kg (range 48 to 89 kg) and mean height was 173.7 cm (range 153 to 188 cm).

Responses to changes in cardiac preloading (first set of measurements) were consistent in all volunteers (Fig. 2, Online Supplementary Table 1, https://links.lww.com/EJA/A373). Increased preloading led to significant increases in SVI and CI [>16% during change from seated (30°) to supine position and more than 35% with passive leg raising]. The increases were accompanied by a nonsignificant decrease in HR and a significant decrease in the TPRI, which led to a small but significant decrease in MAP [9% with the change to supine position (P = 0.001) and 0.8% with passive leg rising (P = 0.67)]. Nonsignificant changes in rSO2 in either hemisphere or in Vmax or Vmean flow velocities were observed.

Fig. 2
Fig. 2:
Variables recorded during the cardiac preload manoeuvres. Values are expressed as mean and SD before and after the test. P values from the generalised estimating equation including time point (pre or post) and baseline value as covariables.

With regard to the responses to gravitational effects, all baseline haemodynamic values in the second set of tests were as expected (Fig. 3, Online Supplementary Table 2, https://links.lww.com/EJA/A373). The gravitational effects in supine, seated and standing positions produced significant differences between positions. Decreases in the SVI exceeded 20%, as did the CI changes between seated – supine and supine – standing positions. We also recorded significant increases in MAP, TPRI and HR.

Fig. 3
Fig. 3:
Responses recorded at baseline between supine decubitus, seated and standing positions. Values are expressed as mean and SD between the different positions. P values from the generalised estimating equation. MCA, middle cerebral artery; rSO2, cerebral regional haemoglobin oxygen saturation.

Absolute cerebral rSO2 values decreased significantly in the right hemisphere but not in the left. The decrease was more consistent between supine and standing positions. We also found significant decreases in absolute rSO2 readings at the deltoids and quadriceps muscles in the different positions. The maximum drop was detected in the quadriceps values from supine to standing.

MCA flow velocities in our study did not behave in a uniform manner in response to position changes from supine to standing. Vmean decreased significantly only when volunteers shifted from the supine to the seated position.

Haemodynamic, rSO2, MCA blood flow velocity and TcPCO2 responses to gravitational effects in the presence of cerebral vasoconstriction and vasodilatation in the three positions during the three experimental ventilatory conditions are summarised in Fig. 4 and Online Supplementary Table 3, https://links.lww.com/EJA/A373. Respiratory responses are shown in Online Supplementary Table 4, https://links.lww.com/EJA/A373. FeO2 values under hyperoxia exceeded the 70% required to establish the experimental condition of cerebral vasoconstriction in all volunteers. Under this condition, CI and HR decreased significantly with all position changes, and MAP increased. Vmean values decreased significantly in all three positions. Vmax values were lower when the volunteers were supine or standing than when they were seated. Cerebral rSO2 increased significantly and to a proportionately greater degree than the increase at the deltoids and quadriceps muscles.

Fig. 4
Fig. 4:
Haemodynamic responses in three different positions. Values are expressed as mean and SD before and after the adoption of each position in each ventilatory condition. P values from the generalised estimating equation including position, time point (pre or post) and baseline value as covariables. MAP, mean arterial pressure; rSO2, cerebral regional haemoglobin oxygen saturation; V mean, mean middle cerebral artery blood flow velocity.

With hypocapnia, ETCO2 decreased between 2.13 and 2.27 kPa (16 to 17 mmHg) with the hyperventilation manoeuvres. CI, HR and MAP increased in all positions whereas Vmax and Vmean decreased in all positions. We observed a slight decrease in both right and left cerebral rSO2 and increases in rSO2 at the deltoid and quadriceps muscles.

The hypercapnia manoeuvre increased ETCO2 values by 4.7 to 6.0 kPa; this was not associated with any significant changes in CI, but HR and MAP increased significantly. MCA Vmax and Vmean flow velocities increased more than 33% from basal values with all position changes and were associated with more marked increases in cerebral rSO2 values than deltoid and quadriceps rSO2 readings.

Transcutaneous CO2 changes in response to the different manoeuvres were significant during hypocapnia and hypercapnia, with lower absolute values than ETCO2 (Online Supplementary Tables 3 and 4, https://links.lww.com/EJA/A373).

All the volunteers were able to maintain an EEP over 6.7 kPa for at least 20 s, reaching the condition for uniform performance of the Valsalva manoeuvre. Haemodynamics, rSO2 and MCA blood flow velocity during the Valsalva manoeuvre in all positions are shown in Fig. 5 and Online Supplementary Table 5, https://links.lww.com/EJA/A373. The CI tended to decrease, although not significantly, in the supine and seated positions but did not change when the volunteers were standing. Cerebral rSO2 changed very little, although there was a significant 3% decrease in the right hemisphere when the volunteers stood. Peripheral rSO2 increased slightly at the deltoid level and decreased significantly at the quadriceps when the manoeuvre was done in the standing position. MCA Vmax decreased significantly during the manoeuvre in all positions while changes in Vmean were NS.

Fig. 5
Fig. 5:
Cerebral and peripheral regional haemoglobin oxygen saturation and maximal middle cerebral artery blood flow velocity during the Valsalva manoeuvre in supine, seated and standing positions. Values are expressed as mean and SD before and after the adoption of each position in each Valsalva manoeuvre. P values from generalised estimating equation including position, time point (pre or post) and baseline value as covariables. rSO2, regional haemoglobin oxygen saturation; V max, maximal middle cerebral artery blood flow velocity.

Discussion

In all our healthy volunteers, brain physiological responses to both gravitational and respiratory challenges were detected by NIRS and transcranial Doppler monitoring. As far as we know, this is the first study applying these combined variables, which may occur during surgical procedures, to the same individuals. Moreover, we could measure CI, adding valuable clinical information on its role to regulate brain perfusion.

Preloading manoeuvres produced a significant increase in CI, confirming the stability of their autonomic nervous system function without changes in MCA flow values (Fig. 2). Contrarily, Ogoh et al.20 showed a lineal relationship between the changes in MCA Vmean and CO. Their external challenges may have produced a deeper effect on carotid-cardiac baroreflex function than our volume redistribution manoeuvre. TPRI and MAP decreased significantly during the orthostatic challenge, but cerebral rSO2 remained steady, probably due to a normal cerebral autoregulation response in our participants.

Changes in body position also altered haemodynamics. CI values decreased but TPRI, MAP and HR increased when volunteers shifted from the supine to the seated position. MCA Vmax was preserved, but Vmean decreased significantly from supine to seated (Fig. 3).

Systemic and cerebral haemodynamic responses can be blunted during anaesthesia and surgery. Jeong et al.21 reported a significant fall in rSO2 greater than 20% in patients undergoing shoulder surgery in the beach-chair position. Other similar trials have found that fewer patients showed a reduction in rSO2 when a sequential compression device was placed on the legs before induction of anaesthesia.22,23 Closhen et al.24 observed a significant drop of about 10% in rSO2 in anaesthetised patients in the beach-chair position, despite actively keeping MAP more than 70 mmHg, and it was immediately restored after repositioning.24 In their control group of volunteers, MAP increased but they did not experience changes in HR or rSO2 from supine. We found in our awake participants a decrease in CI and an increase in MAP from supine to the beach-chair position, but HR increased and there was a decrease in rSO2 in the right hemisphere. Our results support the importance of maintaining an adequate CI during operations performed in the beach-chair position, which has been clearly related to CBF changes, because relying only on MAP may not preclude brain hypoperfusion.

During phase IIb of the Valsalva manoeuvre, the small magnitude of the decreases in rSO2 contrasted with more than 10% decreases in MCA flow velocities. If we had analysed at an earlier point in the manoeuvre (e.g. IIa), decreases would probably have been greater. Our observations are consistent with those of Tiecks et al.,5 who found that flow decreased less in phase IIb than in phase IIa.

We found a greater preservation of rSO2 values in the left hemisphere, suggesting that the dominant hemisphere is more resistant to gravitational effect variations, consistent with other studies.25 In patients undergoing knee replacement, Salazar et al.26 found that the asymmetry in rSO2 values, right lower than left, can warn of postoperative onset of memory decline.26

Normocapnic hyperoxia induced a decrease in CI, a decrease in HR and an increase in MAP, consistent with the known effect of systemic hyperoxia vasoconstriction.27 In this situation, rSO2 values increased while flow velocities fell (Fig. 4). Others have reported CBF decreases attributable to simultaneous decreases in ETCO2.28 Decreases in flow velocities would be expected to cause a decrease in rSO2, but Tisdall et al.14 described the opposite effect, as we did. There is an undervalued risk in multiple situations of anaesthesia and critical care where, in addition to performing pre-oxygenation techniques (hyperoxia), an inadvertent component of hyperventilation with hypocapnia is added, which could increase the risk of cerebral hypoperfusion, mainly in patients at risk (e.g. carotid occlusions).

Both cerebral oximetry and transcranial Doppler are useful as indirect measurement tools for CBF, but in the case of hyperoxia, while transcranial Doppler reflects hypoperfusion due to possible cerebral vasoconstriction, rSO2 is increased by the direct increase of oxygen, in spite of the reduction of blood flow.

Hypocapnia is known to cause cerebral vasoconstriction.29 The MCA Vmean decrease exceeded 40% in all positions studied, and the decrease in rSO2 was about 2 to 4.4% from basal values (Fig. 4). Reductions were similar to those reported in anaesthetised individuals in the beach-chair position and in seated patients with autonomic dysfunction.9,14,30 The hypocapnia test, which required the volunteer to make an inspiratory effort to hyperventilate, was associated with an increase in CI and higher peripheral rSO2 readings. Hypocapnia in an anaesthetised patient under mechanical ventilation might compromise CBF in the absence of rSO2 monitoring given the associated decrease in CI. Alexander et al.,31 who induced ventilatory changes under anaesthesia, reported that hypocapnia could be detected by means of a significant increase in MAP and a decrease in CI. Coverdale et al.32 studied CBF by means of MRI in volunteers under moderate hypocapnia and compared the findings with transcranial Doppler results. They observed high between-subject variability in the constrictor response to hypocapnia, and concluded that alterations in CBF due to manipulations of the PaCO2 are not solely due to changes in the smaller arterioles.

Hypercapnia led to large increases in MCA flow velocities and more marked changes in cerebral rSO2 than any other challenge in our study, independently of gravitational effects due to positioning (Fig. 4). It seems that even healthy individuals with preserved cerebral autoregulation are defenceless when CBF changes arise from hypercapnia. Coverdale et al.32 provoked hypercapnia and found increases in transcranial Doppler but not as important as ours, probably because of differences in ETCO2 values in their study (1.3 kPa) compared to ours (sharp increase between 4.7 and 5.9 kPa). Under conditions of normal ETCO2, a change in MAP should have minimal effect on CBF due to intact cerebral autoregulation. However, under hypercapnic conditions, dynamic cerebral autoregulation is impaired, making it possible that the cerebral circulation responds passively to small MAP changes.32

Our results corroborate the critical direct effect of CO2 partial pressure on brain vasculature. A reflex increase in MAP secondary to hypocapnia could not prevent a drop in CBF. Furthermore, the sudden hypercapnia quickly impaired cerebral autoregulation in our healthy participants.

One limitation of our study was the indirect estimation of CBF by rSO2 and transcranial Doppler. Differences in the angle of insonation, and constriction or stenosis in the vessel being monitored can lead to increases in velocities that are unrelated to CBF. We think that the effects of CO2 on cerebral vessels and CBF are consistent with the responses measured with NIRS and transcranial Doppler ultrasound. However, a key assumption in NIRS technology is that the ratio of arterial to venous blood remains constant, whereas conditions affecting CBF, as in our study, may alter the accuracy of results and is a known limitation. Induction of hypocapnia has been shown to increase the bias in oximeters, overestimating brain tissue oxygenation; with hyperoxia, an underestimation has been found.15 Another limitation may be related to the fact that we measured noninvasive arterial BP with a cuff in the upper arm, instead of a continuous invasive arterial measurement, which would not have been acceptable in our healthy volunteers.

Finally, we must emphasise that the conditions applied in this study were of limited duration in healthy, awake volunteers. Therefore, care must be taken when drawing inferences about effects during longer situations experienced by patients with compromised cerebral autoregulation undergoing surgery under general anaesthesia, or under the effects of sedatives in a critical care unit.

In conclusion, we were able to use transcranial Doppler ultrasound and NIRS to evaluate CBF velocities and oxygenation responses to changes in cardiac preloading (orthostatic tests, passive leg raising, the Valsalva manoeuvre) in healthy volunteers in both physiological and experimental conditions of hyperoxia and hypercapnia and hypocapnia to model cerebral vasoconstriction or vasodilatation. Acute hypercapnia had a greater effect on recorded brain parameters than hypocapnia. The monitoring approach with simultaneously different body position and ventilatory challenges supports the ongoing research of rSO2 feasibility and usefulness as a noninvasive suitable surrogate of CBF during anaesthesia.33–35

Acknowledgements relating to this article

Assistance with the study: we thank Mary Ellen Kerans (independent translation and editing professional) who translated a version of the article.

Financial support and sponsorship: the current study was supported in full by a Spanish National Research Grant (FIS PS09/01603).

Conflicts of interest: none.

Presentation: preliminary data from this study were presented as a poster presentation at the Euroanaesthesia meeting, 2011 in Amsterdam and in the 2011 American Society of Anesthesiologists (ASA) Annual Meeting, Chicago. 15 to 19 October 2011.

References

1. Drummond JC. Blood pressure and the brain: how low can you go? Anesth Analg 2019; 128:759–771.
2. Bijker JB, Persoon S, Peelen LM, et al. Intraoperative hypotension and perioperative ischemic stroke after general surgery: a nested case–control study. Anesthesiology 2012; 116:658–664.
3. Joshi B, Ono M, Brown C, et al. Predicting the limits of cerebral autoregulation during cardiopulmonary bypass. Anesth Analg 2012; 114:503–510.
4. Aaslid R, Lindegaard KF, Sorteberg W, et al. Cerebral autoregulation dynamics in 487 humans. Stroke 1989; 20:45–52.
5. Tiecks FP, Lam AM, Matta BF, et al. Effects of the Valsalva maneuver on cerebral circulation in healthy adults. A transcranial Doppler study. Stroke 1995; 26:1386–1392.
6. Tzeng YC, Ainslie PN. Blood pressure regulation IX: cerebral autoregulation under blood pressure challenges. Eur J Appl Physiol 2014; 114:545–559.
7. Ito H, Kanno I, Fukuda H. Human cerebral circulation: positron emission tomography studies. Ann Nucl Med 2005; 19:65–74.
8. Curley G, Kavanagh BP, Laffey JG. Hypocapnia and the injured brain: more harm than benefit. Crit Care Med 2010; 38:1348–1359.
9. Meng L, Mantulin WW, Alexander BS, et al. Head-up tilt and hyperventilation produce similar changes in cerebral oxygenation and blood volume: an observational comparison study using frequency-domain near-infrared spectroscopy. Can J Anaesth 2012; 59:357–365.
10. Rosner MJ, Coley IB. Cerebral perfusion pressure, intracranial pressure, and head elevation. J Neurosurg 1986; 65:636–641.
11. Pohl A, Cullen DJ. Cerebral ischemia during shoulder surgery in the upright position: a case series. J Clin Anesth 2005; 17:463–469.
12. Calderon-Arnulphi M, Alaraj A, Amin-Hanjani S, et al. Detection of cerebral ischemia in neurovascular surgery using quantitative frequency-domain near-infrared spectroscopy. J Neurosurg 2007; 106:283–290.
13. Gupta AK, Menon DK, Czosnyka M, et al. Noninvasive measurement of cerebral blood volume in volunteers. Br J Anaesth 1997; 78:39–43.
14. Tisdall MM, Taylor C, Tachtsidis I, et al. The effect on cerebral tissue oxygenation index of changes in the concentrations of inspired oxygen and end-tidal carbon dioxide in healthy adult volunteers. Anesth Analg 2009; 109:906–913.
15. Schober A, Feiner JR, Bickler PH, et al. Effects of changes in arterial carbon dioxide and oxygen partial pressures on cerebral oximeter performance. Anesthesiology 2018; 128:97–108.
16. Moritz S, Kasprzak P, Arlt M, et al. Accuracy of cerebral monitoring in detecting cerebral ischemia during carotid endarterectomy: a comparison of transcranial Doppler sonography, near-infrared spectroscopy, stump pressure, and somatosensory evoked potentials. Anesthesiology 2007; 107:563–569.
17. Watzman HM, Kurth CD, Montenegro LM, et al. Arterial and venous contributions to near-infrared cerebral oximetry. Anesthesiology 2000; 93:947–953.
18. Benomar B, Ouattara A, Estagnasie P, et al. Fluid responsiveness predicted by noninvasive bioreactance-based passive leg raise test. Intensive Care Med 2010; 36:1875–1881.
19. Dawson SL, Panerai RB, Potter JF. Critical closing pressure explains cerebral hemodynamics during the Valsalva manoeuvre. J Appl Physiol 1999; 86:675–680.
20. Ogoh S, Brothers RM, Barnes Q, et al. The effect of changes in cardiac output on middle cerebral artery mean blood velocity at rest and during exercise. J Physiol 2005; 569:697–704.
21. Jeong H, Jeong S, Lim HJ, et al. Cerebral oxygen saturation measured by near-infrared spectroscopy and jugular venous bulb oxygen saturation during arthroscopic shoulder surgery in beach chair position under sevoflurane-nitrous oxide or propofol-remifentanil anesthesia. Anesthesiology 2012; 116:1047–1056.
22. Kwak HJ, Lee D, Lee YW, et al. The intermittent sequential compression device on the lower extremities attenuates the decrease in regional cerebral oxygen saturation during sitting position under sevoflurane anesthesia. J Neurosurg Anesthesiol 2011; 23:1–5.
23. Kwak HJ, Lee JS, Lee DC, et al. The effect of a sequential compression device on hemodynamics in arthroscopic shoulder surgery using beach-chair position. Arthroscopy 2010; 26:729–733.
24. Closhen D, Berres M, Werner C, et al. Influence of beach chair position on cerebral oxygen saturation: a comparison of INVOS and FORE-SIGHT cerebral oximeter. J Neurosurg Anesthesiol 2013; 25:414–419.
25. Shinoura N, Yamada R. Dizziness is associated with decreased vasoreactivity in right cerebral hemisphere for head-down manoeuvre – near-infrared spectroscopy study. Clin Physiol Funct Imaging 2005; 25:16–19.
26. Salazar F, Doñate M, Boget T, et al. Relationship between intraoperative regional cerebral oxygen saturation trends and cognitive decline after total knee replacement: a posthoc analysis. BMC Anesthesiol 2014; 14:58.
27. Harten JM, Anderson KJ, Angerson WJ, et al. The effect of normobaric hyperoxia on cardiac index in healthy awake volunteers. Anaesthesia 2003; 58:885–888.
28. Watson NA, Beards SC, Altaf N, et al. The effect of hyperoxia on cerebral blood flow: a study in healthy volunteers using magnetic resonance phase-contrast angiography. Eur J Anaesthesiol 2000; 17:152–159.
29. Ainslie PN, Duffin J. Integration of cerebrovascular CO2 reactivity and chemoreflex control of breathing: mechanisms of regulation, measurement, and interpretation. Am J Physiol Regul Integr Comp Physiol 2009; 296:R1473–R1495.
30. Hunt K, Tachtsidis I, Bleasdale-Barr K, et al. Changes in cerebral oxygenation and haemodynamics during postural blood pressure changes in patients with autonomic failure. Physiol Meas 2006; 27:777–785.
31. Alexander BS, Gelb AW, Mantulin WW, et al. Impact of stepwise hyperventilation on cerebral tissue oxygen saturation in anesthetized patients: a mechanistic study. Acta Anaesthesiol Scand 2013; 57:604–612.
32. Coverdale NS, Gati JS, Oksana Opalevych O, et al. Cerebral blood flow velocity underestimates cerebral blood flow during modest hypercapnia and hypocapnia. J Appl Physiol 2014; 117:1090–1096.
33. Tosh W, Patteril M. Cerebral oximetry. BJA Educ 2016; 16:417–421.
34. Deschamps A, Hall R, Grocott H, et al. Canadian Perioperative Anesthesia Clinical Trials Group. Cerebral oximetry monitoring to maintain normal cerebral oxygen saturation during high-risk cardiac surgery: a randomized controlled feasibility trial. Anesthesiology 2016; 124:826–836.
35. Murphy GS, Greenberg SB, Szokol JW. Safety of beach chair position shoulder surgery: a review of the current literature. Anesth Analg 2019; 129:101–118.

Supplemental Digital Content

Copyright © 2020 European Society of Anaesthesiology. All rights reserved.