To prevent a reduction in cerebral blood flow (CBF), the vasopressor phenylephrine, a selective α-adrenergic drug, is used frequently to treat anesthesia-induced hypotension.1 However, spatially resolved near-infrared spectroscopy-determined (SR-NIRS, i.e., the NIRS optodes are configured to favor measurement of signals from “deep” tissue) frontal lobe tissue oxygenation (ScO2) is reduced with IV phenylephrine infusion despite the increase in arterial pressure,1–4 potentially indicating that phenylephrine induces cerebral ischemia.5 Yet internal carotid artery (ICA) blood flow is unaffected by phenylephrine infusion at a stable arterial partial pressure of oxygen (PaO2) and partial pressure of carbon dioxide (PaCO2),6 indicating that the reduction in ScO2 cannot be attributed to a reduction in cerebral oxygen delivery.
One explanation for the discrepancy between the effect of phenylephrine on ScO2 and ICA blood flow could be that phenylephrine reduces forehead skin blood flow (SkBF), thereby SR-NIRS. Near-infrared (NIR) light is absorbed by hemoglobin in small arteries, capillaries, and venules under the optodes.7 Thus, forehead SkBF may affect the SR-NIRS-derived ScO2, for example, when SkBF increases8–10 or decreases.11 Similarly, Sørensen et al.9 indicate that norepinephrine-induced reduction in ScO2 is influenced by cutaneous vasoconstriction with no change in CBF, but the mechanism that reduces the SR-NIRS-derived ScO2 during phenylephrine infusion remains unclear.
To investigate whether phenylephrine affects ScO2 by way of a reduction in extracranial blood flow as evaluated by flow in the external carotid artery (ECA), we measured forehead SkBF, ICA, and ECA blood flow besides flow in the vertebral artery (VA) by duplex ultrasonography together with ScO2 and middle cerebral artery mean blood velocity (MCA Vmean) during IV infusion of phenylephrine. We hypothesized that phenylephrine would mediate a reduction in ScO2 by attenuated ECA blood flow and thus SkBF, while ICA and VA blood flow, as MCA Vmean would remain unaffected.
The study was approved by the Regional Ethics Committee (H-4-2010-132) in accordance with the Declaration of Helsinki. Seven healthy men volunteers (25  years, 182  cm, and 77  kg; mean [SD]) participated in the study after providing written informed consent after explanation of the procedures and their potential risks. The subjects were free from overt cardiovascular, pulmonary, metabolic, or neurological diseases and did not take any medication. All subjects were sedentary to moderately active, nonsmokers and were requested to abstain from caffeinated beverages for 12 hours and strenuous physical activity and alcohol for 24 hours before experimental sessions. On experimental days, the subjects arrived at the laboratory a minimum of 2 hours after a light meal and were familiarized with the equipment and procedures before any measurements were made.
Arterial pressure was measured through a transducer (Edwards Life Science, Irvine, CA) placed at heart level from a 1.1-mm catheter in the brachial artery of the nondominant arm, and mean arterial blood pressure (MAP) was expressed by integration of arterial pressure waveform (Danica, Copenhagen, Denmark). Since the subjects were supine, MAP in the brachial artery was taken to represent that in the arteries serving the brain. A second catheter (0.7 mm) was inserted in the right subclavian vein through an arm vein for the administration of phenylephrine. Right MCA Vmean was measured by using transcranial Doppler ultrasound (DWL, Sipplingen, Germany) with a 2-MHz probe placed over the temporal ultrasound window and fixed with an adjustable headband and adhesive ultrasonic gel (Tensive, Parker Laboratories, Orange, NJ). Frontal lobe ScO2 was monitored by SR-NIRS (INVOS-4100 Cerebral Oximeter, Covidien, Mansfield, MA), while for SkBF laser Doppler, flowmetry with 785-nm light was used (VMS-LDF, Moor Instruments, Axminster, UK). To avoid influence from supraorbital cutaneous blood flow, the supraorbital sinus and frontal vein, the sensors were placed above the supraorbital edge and lateral on the forehead.12,13 Data were analog-digital converted, sampled at 1 kHz (Powerlab, ADInstruments, Colorado Springs, CO), and stored for later analysis (Labchart, ADInstruments).
The right ICA, ECA, and VA blood flows were examined by using duplex Doppler ultrasonography (Vivid-e, GE Healthcare, Tokyo, Japan). For blood flow measurements, we used the brightness mode to determine mean vessel diameter in a longitudinal section. ICA measurements were performed approximately 1.0 to 1.5 cm distal to the carotid bifurcation, while the subject’s chin was slightly elevated. ECA measurements were also obtained approximately 1.0 to 1.5 cm above the carotid bifurcation, while VA blood flow was measured between the transverse processes of the C3 and the subclavian artery. The Doppler velocity spectrum was subsequently identified by pulsed wave mode. The systolic and diastolic diameters were measured, and the mean diameter was taken as [(systolic diameter × 1/3)] + [(diastolic diameter × 2/3)]. The time-averaged mean flow velocity obtained in pulsed wave mode was measured by tracing the average flow rate for each time phase and by calculating the time-averaged value across approximately 45 cardiac cycles to eliminate oscillatory effects caused by ventilation. When recording blood flow velocity, care was taken to ensure that the probe position was stable, that the insonation angle did not vary (< 60° in most cases), and that the sample volume was positioned in the center of the vessel and adjusted to cover the width of the vessel. The mean blood flow velocity was calculated on the basis of velocity waveforms traced by the apparatus software. Finally, blood flow was calculated by multiplying the cross-sectional area [π × (mean diameter/2)2] with mean blood flow velocity: Blood flow (mL/min) = 60 × mean blood flow velocity (cm/s) × area (cm2).14 Vascular conductance in each of the arteries, intraextracranial vascular resistance, and a middle cerebral artery conductance index were calculated as:
Cerebral artery conductance = (ICA, ECA, or VA) blood flow/MAP
Intracranial vascular resistance = MAP/(ICA + VA blood flow)
Extracranial vascular resistance = MAP/ECA blood flow Middle cerebral artery conductance index = MCA Vmean/MAP
Because only 1 Doppler machine was used, blood flow velocity and diameters of ICA, VA, and ECA were recorded for at least 1 minute each from the steady-state period (10–15th minutes after the start of the phenylephrine infusion) in random order (Fig. 1). Other continuously measured variables (heart rate, MAP, SkBF, MCA Vmean, and ScO2) were calculated as the average at the end of baseline and for phenylephrine infusion between the 14th and 15th minutes.
An arterial blood sample was obtained at rest and 8 to 10 minutes after the start of the phenylephrine infusion and immediately analyzed for PaCO2, PaO2, and pH by using an ABL 725 apparatus (Radiometer, Copenhagen, Denmark).
After catheterization and instrumentation, the subjects rested supine for 30 minutes with an elevated headrest. The experiment was conducted in randomized order with 30 minutes between control and phenylephrine trials, and the subjects acted as their own controls to minimize the number of subjects included in this invasive study. Phenylephrine infusion was increased until MAP was 30 mm·Hg above baseline. The phenylephrine infusion was at 1.66 ± 0.08 μg/kg/min, and the infusion time was 15 minutes.
To determine the number of subjects needed for the study, we performed a power test by using the ScO2 data (an anticipated 10.5% changes in ScO2) from our previous study.6 In the power paired t test (δ = 10.5, SD = 7.6, σ level = 0.05, power = 0.8) calculation, sample size should be n >6 to obtain a significant difference. The difference in steady-state hemodynamic responses between the 2 conditions (baseline and phenylephrine infusion) was assessed by the Wilcoxon-Mann-Whitney test because Wilcoxon signed rank test (pairwise analysis) is generally not applicable to a before/after design.15,16
In addition, the relationships between baseline value and changes in SkBF and ECA blood flow after phenylephrine infusion were evaluated by Spearman rank-order correlation, to confirm a difference between baseline and phenylephrine conditions. Correlations between changes in ScO2 and SkBF or cerebral vascular conductance for the ECA after phenylephrine infusion from baseline were evaluated by both Spearman rank-order correlation and linear regression analysis. The 95% confidence interval (CI) was calculated as
that follows the t distribution with degrees of freedom n − 2. A P value of <0.05 was considered to indicate a statistically significant difference. Data are ex pressed as means ± SE, and analysis was conducted by using SigmaStat (Jandel Scientific Software; SPSS; Chicago, IL).
During phenylephrine infusion, MAP increased, while heart rate decreased (Table 1). MCA Vmean was unchanged, while ScO2 decreased during the infusion of phenylephrine with no significant change in PaCO2, PaO2, or pH. There was no significant difference in SkBF (P = 0.32) between rest and phenylephrine infusion because of a large variation in baseline values.
ICA, ECA, and VA Diameter, Velocity, and Blood Flow
Administration of phenylephrine left ICA and VA blood flow unchanged with no significant change in their diameter despite increase in their mean blood velocity (Table 1 and Fig. 2). In contrast, administration of phenylephrine may have reduced ECA blood flow (P = 0.073) and also that response was because of a decrease in mean blood velocity at an unchanged diameter. In addition, the rank correlation between baseline values and changes in ECA blood flow during phenylephrine infusion was significant (P < 0.0001, Fig. 3). This finding indicates that ECA blood flow decreased during phenylephrine infusion.
Cerebral Blood Flow Variables Normalized by Perfusion Pressure and Relationship Between ScO2 and ECA Conductance
During administration of phenylephrine, VA, ICA conductance, and MCA conductance index were unchanged, while ECA conductance may have decreased (P = 0.073, Table 2). The decrease in extracranial vascular conductance (ECA) was much larger than that in intracranial vasculature (ICA + VA) (% changes in ECA vs ICA + VA; −47% ± 4% vs −17% ± 4%, respectively; P < 0.001). Spearman correlation coefficient between ScO2 and forehead SkBF and ECA conductance was 0.81 (P < 0.001, 95% CI, 0.49–0.94) and 0.64 (P = 0.012, 95% CI, 0.17–0.88), respectively (Spearman correlation coefficients between forehead SkBF and ECA conductance was 0.53; P = 0.047, 95% CI, 0.003–0.83). By linear regression analysis, the relationship between ScO2 and forehead SkBF indicated a Pearson r value of 0.55 (P = 0.042, 95% CI, 0.025–0.84), and for ECA conductance, the Pearson r value was 0.62 (P = 0.019, 95% CI, 0.13–0.86; Fig. 4).
The main finding of the present study is that spatially resolved NIR spectroscopy evaluation of cerebral tissue oxygenation does not account for an extracerebral influence when SkBF is reduced by phenylephrine administration. Thus, phenylephrine, a selective α-agonist, causes vasoconstriction in the extracerebral vasculature but not in the cerebral vasculature, and yet ScO2 decreases. In this study, both ICA and VA blood flow remained unchanged in response to phenylephrine as did MCA Vmean. In contrast, blood flow in the ECA that serves the dominant part of the extracerebral tissue of the head was reduced during phenylephrine infusion as was forehead SkBF and the SR-NIRS-determined ScO2. Moreover, there was a correlation between the reduction in the SR-NIRS-determined frontal lobe tissue oxygenation and that in forehead SkBF and ECA conductance during the infusion of phenylephrine. Together these findings suggest that a phenylephrine-induced decrease in ScO2, as determined by SR-NIRS, is influenced by vasoconstriction in the extracranial vasculature rather than a reduction in cerebral oxygenation.
The cerebral arteries are richly innervated with sympathetic nerve fibers,17,18 and there is a potential for α-adrenergic modulation of the cerebral vasculature.19–21 However, the cerebral circulation is protected from circulating catecholamines by the blood–brain barrier. Indeed, phenylephrine does not cause cerebral vasoconstriction even during canine cardiopulmonary bypass, which may alter the permeability of the blood–brain barrier.22 Here we support these findings in healthy young volunteers. Yet although there was no significant change in ICA and VA blood flow or conductance, the intracranial vascular (ICA + VA) resistance tended to increase (Table 2, P = 0.097). We consider this reduction in cerebrovascular conductance an autoregulatory effect secondary to the hypertension induced by phenylephrine, although a direct effect of phenylephrine on the cerebral vessels cannot be excluded.
The use of SR-NIRS to monitor ScO2 during administration of a vasopressor drug is of clinical importance. During anesthesia-induced hypotension, infusion of vasopressor drugs is used to increase arterial blood pressure to maintain cerebral perfusion and potentially improve postoperative outcome.23,24 Nevertheless, the decrease in SR-NIRS-determined ScO2 during administrations of vasopressors could illustrate an inability of NIRS, even with spatial resolution, to adequately suppress light returning from extracranial tissue including the skin.9,11 Our group6 demonstrated that administration of phenylephrine left ICA blood flow unaffected, suggesting that cerebral oxygenation was not reduced despite a decrease in SR-NIRS-determined ScO2. In addition, phenylephrine infusion led to a decrease in extracerebral blood flow. With a higher phenylephrine dosage in this study compared with the previous study,6 MAP was also higher (116 vs 97 mm·Hg), but ICA, VA, and MCA Vmean blood flow remained unchanged. In contrast, ECA blood flow decreased, and the decrease was related to the reduction in SkBF. A laser Doppler measurement reflects skin blood velocity rather than blood flow; however, the parallel reduction in ECA blood flow demonstrates that the laser Doppler signal is an index of blood flow during phenylephrine infusion.
Although SR-NIRS-determined cerebral oxygenation is reduced with the administration of phenylephrine, SR-NIRS determined cerebral oxygenation is preserved during administration of ephedrine.1,4 Phenylephrine and ephedrine have different pharmacological profiles25; however, with α-adrenoceptor activation, both reduce SkBF.26,27 To what extent ephedrine affects CBF and SR-NIRS requires clarification. Forehead SkBF could influence the SR-NIRS-determined ScO2 signal because NIR light must penetrate the overlying skin, subcutaneous adipose tissue, and the scalp before it reaches the target tissues, that is, the brain. Thus, an influence of superficial tissues on NIR light absorption and scattering should be considered. Indeed, recent studies9,11 demonstrated that extracerebral contamination significantly affects SR-NIRS measurements of cerebral oxygen saturation. Finally, during carotid endarterectomy, cross-clamping of the common carotid artery reduces ScO2, but ScO2 returns to the control value when flow in the ECA is reestablished.12 Therefore, it is reasonable that changes in SkBF make ScO2 underestimate cerebral oxygenation during phenylephrine infusion as indicated by a linear relationship between changes in ScO2 and forehead SkBF and ECA conductance (Fig. 4). It remains, however, that CBF and cerebral oxygenation may be influenced by, for example, perfusion pressure, cardiac output, and sympathetic nerve activity.28–30
This study did not establish a dose–effect relationship about the effect of phenylephrine on intracerebral versus extracerebral tissue flow, conductance, or oxygenation. We made a comparison to the data presented by Ogoh et al.6 regarding intracerebral flow for the ICA during which a lower dosage of phenylephrine was used. The change in ICA blood flow (by +17%) with the administration of phenylephrine was much larger than for the VA (+7%), and further regional differences in CBF responses to phenylephrine could be evaluated by magnetic resonance spectroscopy. Similarly, regional differences in α-receptor density need to be addressed. For the main focus of the article addressing a possible influence of SkBF on ScO2, we established a relationship between the interindividual reduction in ScO2 and reductions in both SkBF and ECA, supporting that oxygenation of extracranial tissue plays a role in the reduction in ScO2 provoked by the administration of phenylephrine. However, we cannot exclude a change in cerebral blood volume composition, allowing more venous blood to contribute to the reduction in ScO2 on administration of phenylephrine. Furthermore, we interpreted changes in VA and ICA blood flow in response to phenylephrine to be equal on the right and left side of the brain. Last, the effect of phenylephrine infusion on ScO2 may be device-dependent.11 We accept that the Wilcoxon signed rank test is generally not ideal for a before/after design.15,16 Typically in a pairwise analysis, the error term is correlated to the baseline and that was the case for both SkBF and ECA blood flow; in that, the lowest baseline had the smallest decreases and vice-versa with administration of phenylephrine (Fig. 2). However, considering that both variables decreased in all subjects with administration of phenylephrine, the probability of that finding is 0.5 times the number of subjects (7), that is, P = 0.008 not appreciating the individual differences.
In summary, phenylephrine infusion reduces extracranial blood flow, while CBF remains unchanged. Thus, an effect of phenylephrine on external carotid artery blood flow, as supported by reduced skin blood flow, rather than change in cerebral oxygenation affects the INVOS-4100 SR-NIRS determined by frontal lobe tissue oxygenation during infusion of phenylephrine.
Name: Shigehiko Ogoh, PhD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: S. Ogoh has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Kohei Sato, PhD.
Contribution: This author helped conduct the study and analyze the data.
Attestation: K. Sato has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Kazunobu Okazaki, PhD.
Contribution: This author helped conduct the study.
Attestation: K. Okazaki has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Tadayoshi Miyamoto, PhD.
Contribution: This author helped conduct the study.
Attestation: T. Miyamoto has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Frederik Secher.
Contribution: This author helped conduct the study.
Attestation: F. Secher has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Henrik Sørensen, PhD.
Contribution: This author helped design and conduct the study and analyze the data.
Attestation: H. Sørensen has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Peter Rasmussen, PhD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: P. Rasmussen has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Niels H. Secher, MD, DMSc.
Contribution: This author helped design and conduct the study and write the manuscript.
Attestation: N. H. Secher has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
This manuscript was handled by: Gregory J. Crosby, MD.
We appreciate the time and effort spent by the volunteer subjects in the present study.
1. Nissen P, Brassard P, Jørgensen TB, Secher NH. Phenylephrine but not ephedrine reduces frontal lobe oxygenation following anesthesia-induced hypotension. Neurocrit Care. 2010;12:17–23
2. Brassard P, Seifert T, Secher NH. Is cerebral oxygenation negatively affected by infusion of norepinephrine in healthy subjects? Br J Anaesth. 2009;102:800–5
3. Brassard P, Seifert T, Wissenberg M, Jensen PM, Hansen CK, Secher NH. Phenylephrine decreases frontal lobe oxygenation at rest but not during moderately intense exercise. J Appl Physiol (1985). 2010;108:1472–8
4. Meng L, Cannesson M, Alexander BS, Yu Z, Kain ZN, Cerussi AE, Tromberg BJ, Mantulin WW. Effect of phenylephrine and ephedrine bolus treatment on cerebral oxygenation in anaesthetized patients. Br J Anaesth. 2011;107:209–17
5. Al-Rawi PG, Kirkpatrick PJ. Tissue oxygen index: thresholds for cerebral ischemia using near-infrared spectroscopy. Stroke. 2006;37:2720–5
6. Ogoh S, Sato K, Fisher JP, Seifert T, Overgaard M, Secher NH. The effect of phenylephrine on arterial and venous cerebral blood flow in healthy subjects. Clin Physiol Funct Imaging. 2011;31:445–51
7. Mancini DM, Bolinger L, Li H, Kendrick K, Chance B, Wilson JR. Validation of near-infrared spectroscopy in humans. J Appl Physiol (1985). 1994;77:2740–7
8. Germon TJ, Evans PD, Manara AR, Barnett NJ, Wall P, Nelson RJ. Sensitivity of near infrared spectroscopy to cerebral and extra-cerebral oxygenation changes is determined by emitter-detector separation. J Clin Monit Comput. 1998;14:353–60
9. Sørensen H, Secher NH, Siebenmann C, Nielsen HB, Kohl-Bareis M, Lundby C, Rasmussen P. Cutaneous vasoconstriction affects near-infrared spectroscopy determined cerebral oxygen saturation during administration of norepinephrine. Anesthesiology. 2012;117:263–70
10. Takahashi T, Takikawa Y, Kawagoe R, Shibuya S, Iwano T, Kitazawa S. Influence of skin blood flow on near-infrared spectroscopy signals measured on the forehead during a verbal fluency task. Neuroimage. 2011;57:991–1002
11. Davie SN, Grocott HP. Impact of extracranial contamination on regional cerebral oxygen saturation: a comparison of three cerebral oximetry technologies. Anesthesiology. 2012;116:834–40
12. Hove JD, Rosenberg I, Sejrsen P, Hove KD, Secher NH. Supraorbital cutaneous blood flow rate during carotid endarterectomy. Clin Physiol Funct Imaging. 2006;26:323–7
13. Kishi K, Kawaguchi M, Yoshitani K, Nagahata T, Furuya H. Influence of patient variables and sensor location on regional cerebral oxygen saturation measured by INVOS 4100 near-infrared spectrophotometers. J Neurosurg Anesthesiol. 2003;15:302–6
14. Sato K, Sadamoto T. Different blood flow responses to dynamic exercise between internal carotid and vertebral arteries in women. J Appl Physiol (1985). 2010;109:864–9
15. Diehr P, Martin DC, Koepsell T, Cheadle A. Breaking the matches in a paired t-test for community interventions when the number of pairs is small. Stat Med. 1995;14:1491–504
16. Proschan MA. On the distribution of the unpaired t-statistic with paired data. Stat Med. 1996;15:1059–63
17. Edvinsson L. Neurogenic mechanisms in the cerebrovascular bed. Autonomic nerves, amine receptors and their effects on cerebral blood flow. Acta Physiol Scand Suppl. 1975;427:1–35
18. Nelson E, Rennels M. Innervation of intracranial arteries. Brain. 1970;93:475–90
19. Alm A, Bill A. The effect of stimulation of the cervical sympathetic chain on retinal oxygen tension and on uveal, retinal and cerebral blood flow in cats. Acta Physiol Scand. 1973;88:84–94
20. Mitchell DA, Lambert G, Secher NH, Raven PB, van Lieshout J, Esler MD. Jugular venous overflow of noradrenaline from the brain: a neurochemical indicator of cerebrovascular sympathetic nerve activity in humans. J Physiol. 2009;587:2589–97
21. Ogoh S, Brothers RM, Eubank WL, Raven PB. Autonomic neural control of the cerebral vasculature: acute hypotension. Stroke. 2008;39:1979–87
22. Johnston WE, DeWitt DS, Vinten-Johansen J, Stump DA, Prough DS. Phenylephrine does not reduce cerebral perfusion during canine cardiopulmonary bypass. Anesth Analg. 1994;79:14–8
23. Murkin JM, Adams SJ, Novick RJ, Quantz M, Bainbridge D, Iglesias I, Cleland A, Schaefer B, Irwin B, Fox S. Monitoring brain oxygen saturation during coronary bypass surgery: a randomized, prospective study. Anesth Analg. 2007;104:51–8
24. Slater JP, Guarino T, Stack J, Vinod K, Bustami RT, Brown JM 3rd, Rodriguez AL, Magovern CJ, Zaubler T, Freundlich K, Parr GV. Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg. 2009;87:36–44 discussion 44–5
25. Kobayashi S, Endou M, Sakuraya F, Matsuda N, Zhang XH, Azuma M, Echigo N, Kemmotsu O, Hattori Y, Gando S. The sympathomimetic actions of l-ephedrine and d-pseudoephedrine: direct receptor activation or norepinephrine release? Anesth Analg. 2003;97:1239–45
26. Crandall CG, Shibasaki M, Yen TC. Evidence that the human cutaneous venoarteriolar response is not mediated by adrenergic mechanisms. J Physiol. 2002;538:599–605
27. Lecoq JP, Brichant JF, Lamy ML, Joris JL. Norepinephrine and ephedrine do not counteract the increase in cutaneous microcirculation induced by spinal anaesthesia. Br J Anaesth. 2010;105:214–9
28. Ogoh S, Ainslie PN. Regulatory mechanisms of cerebral blood flow during exercise: new concepts. Exerc Sport Sci Rev. 2009;37:123–9
29. Ogoh S, Ainslie PN. Cerebral blood flow during exercise: mechanisms of regulation. J Appl Physiol (1985). 2009;107:1370–80
30. Ogoh S, Brothers RM, Barnes Q, Eubank WL, Hawkins MN, Purkayastha S, O-Yurvati A, Raven PB. 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