Cerebral and muscle tissue oxygenation in acute hypoxic ventilatory response test

Cerebral and muscle tissue oxygenation in acute hypoxic ventilatory response test

Respiratory Physiology & Neurobiology 155 (2007) 71–81 Cerebral and muscle tissue oxygenation in acute hypoxic ventilatory response test Juha E. Pelt...

323KB Sizes 0 Downloads 50 Views

Respiratory Physiology & Neurobiology 155 (2007) 71–81

Cerebral and muscle tissue oxygenation in acute hypoxic ventilatory response test Juha E. Peltonen a,b,∗ , John M. Kowalchuk b , Donald H. Paterson b , Darren S. DeLorey b,c , Gregory R. duManoir b , Robert J. Petrella b,d , J. Kevin Shoemaker e b

a Unit for Sports and Exercise Medicine, Institute of Clinical Medicine, University of Helsinki, Helsinki 00250, Finland Canadian Centre for Activity and Aging, School of Kinesiology, The University of Western Ontario, London, Ont., Canada N6G 2M3 c Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, WI 53295, USA d Schulich School of Medicine and Dentistry, The University of Western Ontario, London, Ont., Canada N6A 5C1 e Neurovascular Research Laboratory, School of Kinesiology, The University of Western Ontario, London, Ont., Canada N6A 3K7

Accepted 28 March 2006

Abstract Eight men were exposed to progressive isocapnic hypoxia for 10 min to test the hypothesis that (i) cerebral and muscle tissue would follow similar deoxygenation profiles during an acute hypoxic ventilatory response (AHVR) test; and (ii) strong cerebrovascular responsiveness to hypoxia would be related to attenuated cerebral deoxygenation. End-tidal O2 concentration was reduced from normoxia (∼102 mmHg) to ∼45 mmHg while arterial oxygen saturation (SpO2 %) declined from 98 ± 1% to 77 ± 7% (P < 0.001). Near-infrared spectroscopy (NIRS)-derived local cerebral tissue (frontal lobe) deoxyhemoglobin increased 5.55 ± 2.22 ␮M, while oxyhemoglobin and tissue oxygenation index decreased 2.57 ± 1.99 ␮M and 6.2 ± 3.4%, respectively (all P < 0.001). In muscle (m. vastus lateralis) the NIRS changes from the initial normoxic level were non-significant. Cerebral blood velocity (Vmean , transcranial Doppler) in the middle cerebral artery increased from 53.4 ± 10.4 to 60.6 ± 11.6 cm s−1 (P < 0.001). In relation to the decline in SpO2 % the mean rate of increase of Vmean and AHVR were 0.33 ± 0.19 cm s−1 %−1 and 0.52 ± 0.20 l min−1 %−1 , respectively. We conclude that cerebral, but not muscle, tissue shows changes reflecting a greater deoxygenation during acute hypoxia. However, the changes in NIRS parameters were not related to cerebrovascular responsiveness or ventilatory chemosensitivity during graded hypoxia. © 2006 Elsevier B.V. All rights reserved. Keywords: Near-infrared spectroscopy; Cerebral blood flow; Transcranial Doppler; Hypoxia; Acute; Chemosensitivity; Ventilation

1. Introduction Sufficient oxygenation and perfusion of brain tissue is vital for avoidance of hazardous symptoms while at hypoxia/altitude or after effects of chronic hypoxia (Hornbein, 2001). Previous studies indicate that with exposure to acute hypoxia, cerebral blood flow (CBF) (Cohen et al., 1967) and CBF velocity (Jensen et al., 1996; Kolb et al., 2004b) increase to maintain O2 delivery. It has been suggested that cerebral O2 supply is well protected in acute hypoxia (Roach and Hackett, 2001). Thus, whole brain O2 consumption is unaffected even with severe degrees of hypoxemia (Cohen et al., 1967) and the energy state of the brain tissue remains remarkably normal by arterial O2 tensions (PaO2 ) as low as 56 mmHg, although intracellular energy state is compromised



Corresponding author. Tel.: +358 9 43 42 100; fax: +358 9 49 08 09. E-mail address: [email protected] (J.E. Peltonen).

1569-9048/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.resp.2006.03.008

when PaO2 is further reduced to 40 mmHg (Rolett et al., 2000; Raichle and Hornbein, 2001). In contrast, the turnover of several cerebral neurotransmitters, e.g. acetylcholine, is reduced even by mild hypoxia (Gibson et al., 1981; Raichle and Hornbein, 2001). The recent papers, however, indicate a drop in regional cerebral oxygen saturation in acute hypoxia. Kolb et al. (2004b) reported cerebral tissue saturation reduction from an average of 74% to approximately 66% during an acute hypoxic ventilatory response (AHVR) test. In the study of Imray et al. (2003) not only was regional cerebral saturation reduced from 70% at 150 m to 66% after a night at 3459 m but also muscle tissue saturation decreased from 73% to 68%. In the acute ventilatory response to hypoxia the afferent impulses from the carotid body reach the nucleus tractus solitarius leading to the stimulation of ventilation (V˙ E ) (Burton and Kazemi, 2000). The chemosensitivity of this ventilatory response to lowered arterial O2 tensions is monitored with an AHVR test, while end-tidal CO2 tension is clamped to iso-

72

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

capnia (Weil and Zwillich, 1976; Kolb et al., 2004b). Blunted chemosensitivity has been observed in athletes of outstanding aerobic endurance capacity in comparison with normal controls, while experienced climbers show vigorous respiratory responses to hypoxia to maintain adequate oxygenation (Schoene, 1982). Although increased ventilation in hypoxia is beneficial in elevating arterial O2 saturation (Gavin et al., 1998; Katayama et al., 2001) hyperventilation reduces PaCO2 , and this has a powerful vasoconstrictor effect on CBF (Ellingsen et al., 1987; Ainslie and Poulin, 2004). This may cause poorer oxygenation of the brain despite greater ventilation, likely because of a hypocapniainduced decrease in CBF that more than offsets the increase in arterial O2 saturation (Hornbein et al., 1989). At altitude, the end result seems to be that persons with a more vigorous ventilatory response to hypoxia have more neurobehavioral impairment like slowed performance on more complex tasks of cognitive and motor function (Hornbein, 2001). During acute hypoxia, Ainslie and Poulin (2004) have shown the inter-individual variability in AHVR to be firmly linked to the variability in CBF velocity as individuals with a high AHVR were found also to have high CBF velocity response to hypoxia. However, the integrative ventilatory, cerebrovascular and cardiovascular responses to hypoxia were strongly influenced by arterial CO2 tension (PaCO2 ). During poikilocapnia hypoxia, a high AHVR blocked much of the acute hypoxic cerebral blood flow responses due to hypocapnia-induced cerebral vasoconstriction. Conversely, during hypercapnic hypoxia, a high AHVR was associated with a high cerebrovascular responsiveness, demonstrating a linkage of individual sensitivities of ventilation and cerebral blood flow to the interaction of PaCO2 and hypoxia. Interestingly, during isocapnia, the strong correlation between ventilatory and cerebrovascular responsiveness was not found (Ainslie and Poulin, 2004). To the best of our knowledge, there are no published studies that have linked ventilatory and cerebrovascular responsiveness to both cerebral and muscle tissue oxygenation and deoxygenation status during acute hypoxia. Therefore, the purpose of the present study was to examine (i) if cerebral and muscle tissue differ in defending against tissue deoxygenation during an acute progressive isocapnic hypoxia, and (ii) the relationship between cerebrovascular responsiveness, ventilatory chemosensitivity and the extent of tissue deoxygenation to graded hypoxia. With an AHVR test that includes clamped end-tidal O2 tension the ventilatory response cannot succeed in adjusting arterial Hb saturation. Thus, this model allowed testing of cardiovascular compensation on tissue deoxygenation. We tested the hypothesis that (i) cerebral and muscle tissue would follow similar deoxygenation profiles during acute isocapnic hypoxia; and (ii) strong cerebrovascular responsiveness to hypoxia would be related to attenuated cerebral deoxygenation. 2. Materials and methods 2.1. Subjects Eight healthy males (185.4 ± 4.8 cm, 78.9 ± 3.0 kg, 28 ± 5 years) volunteered and gave written, informed consent to par-

ticipate in the study. All procedures where approved by The University of Western Ontario Health Sciences Research Ethics Board. All subjects were physically active, ranging from recreational exercisers to competitive cyclists. The subjects were medically screened (including ultrasound study of the heart and the carotid arteries, and a standard 12-lead ECG at rest) and had no history of cardiovascular, respiratory, or musculoskeletal diseases and were free of medication. 2.2. Experimental sequence Participants reported to the laboratory 3–4 h post-meal consumption and after 24 h without caffeine or alcohol ingestion and ≥12 h after physical exercise. After a familiarization period of 5 min while breathing through a mouthpiece with nose occluded the AHVR test was performed while the subject sat quietly and relaxed. The AHVR test started with the subject breathing room air for 2 min. Then, inspired O2 concentration was gradually reduced from normoxia to hypoxia over 10 min while end-tidal CO2 (PETCO2 ) was maintained at a previously determined resting level (the mean resting PETCO2 for each subject was obtained from two or three recent breath-by-breath data that were recorded for other purposes). The hypoxic stimulus was varied by holding the end-tidal O2 (PETO2 ) constant at 10 different predetermined descending steps (PETO2 = 102, 96, 89, 82, 74, 67, 60, 55, 50, 45 mmHg), each step lasting 60 s. The selection of PETO2 was based on previous studies (Weil et al., 1970; Sahn et al., 1977; Harms and Stager, 1995; Kolb et al., 2004b). Terminating AHVR test at or close to PETO2 45 mmHg is a common practice, as this level will induce the steep part of V˙ E versus PETO2 (or versus alveolar O2 tension, PAO2 ) curve yet providing a safety margin (Magosso and Ursino, 2004) for the well-being of the subjects. A 10 min hypoxic period was chosen to avoid possible hypoxic ventilatory decline during longer exposures (Smith et al., 2001). A dynamic end-tidal forcing technique (Poulin et al., 1993; Vovk et al., 2002) was used to clamp PETO2 and PETCO2 at a desired level and to change O2 or CO2 concentrations instantaneously. Accurate control of the end-tidal gases was achieved with a computer-controlled fast gas mixing system as previously described (Poulin et al., 1993; Vovk et al., 2002). Briefly, the controlling computer generated the inspired partial pressures of O2 and CO2 by comparison of the measured end-tidal with the desired values and adjusted the gas mixture for the next inspiration as needed. Immediately after the end of the AHVR test, 100% O2 was delivered to the subjects via respiratory mask to hasten the recovery of SpO2 % to the normal level. 2.3. Measurements of cardiorespiratory responses Heart rate (fH) was continuously monitored by electrocardiogram (ECG) and arterial O2 saturation (SpO2 %) by pulse oximetry (Nonin 8600, Nonin Medical Inc., Plymouth, MN) from the earlobe. Ventilation (V˙ E ) and alveolar gas exchange including endtidal partial pressures for O2 and CO2 were measured breathby-breath throughout the AHVR protocol. The subjects breathed

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

through a mouthpiece connected to a low-deadspace (90 ml) low-resistance turbine (Alpha Technologies VMM 110, Aliso Viejo, CA) for measurement of inspiratory and expiratory volumes and flow. The turbine was calibrated before each test by using a syringe of known volume (3.01 l, Hans Rudolph Inc., Kansas City, MO). A pneumotachograph attached between the volume turbine and a 50 mm diameter inspiratory mixing chamber was used to record breathing phase for the micro computer controlling the dynamic gas forcing system. Inspired and expired gases were sampled continuously at the mouth and analyzed for concentrations of O2 , CO2 , N2 and Ar by mass spectrometry (AMIS 2000; Innovision A/S, Odense, Denmark) after calibration with precision analyzed gas mixtures. Changes in gas concentration were aligned with gas volumes by measuring the time delay for a square-wave bolus of gas passing the turbine to the resulting changes in fractional gas concentrations as measured by the mass spectrometer. Data collected every 20 ms were transferred to a computer, which aligned concentrations with volume data to build a profile of each breath. Breath-by-breath alveolar gas exchange was calculated by using algorithms of Beaver et al. (1981) and interpolated to give values in 1 s intervals. 2.4. Regional cerebral and muscle oxygenation Local cerebral (C) and muscle (M) oxygenation profiles were monitored with two near-infrared spectroscopy (NIRS) (NIRO 300, Hamamatsu Photonics K.K., Hamamatsu City, Japan) systems. For cerebral monitoring the optodes were placed over the right frontal cortex, approximately 2 cm above the right eyebrow as laterally as possible. Thus, the emitting optode was placed approximately 1–2 cm laterally to the sagittal midline of the head and the receiving optode laterally thereof. A 5 cm interoptode distance was used with seven of the eight subjects and a 4 cm distance with one subject. This optode position was selected to maximize interrogation from an area that is supplied primarily by the middle cerebral artery (MCA). However, there is potentially some overlapping within the NIRS signal from those areas that are supplied by the anterior cerebral artery (ACA). For both cerebral and muscle measurements, the optodes were housed in an optically dense plastic holder to minimize the intrusion of extraneous light and loss of transmitted NIR light from the field of interrogation and to ensure that the position of the optodes, relative to each other, was fixed and invariant. The optode assembly was secured on the head with a two-sided tape and then covered with a black sweatband. For local muscle oxygenation profiles, the transmitting and receiving optodes were placed on the vastus lateralis muscle of the right leg, at mid thigh level and parallel with the long axis of the muscle at 5 cm interoptode distance. The optode holder was attached to the skin with tape and covered with an elastic cloth. The theory of NIRS is described in detail elsewhere (Elwell, 1995). Briefly, one fiber optic bundle carried the NIR light produced by laser diodes to the tissue of interest while a second optic fiber bundle returned the transmitted light from the tissue to a photon detector (photomultiplier tube, PMT) in the spec-

73

trometer. Four different wavelength laser diodes (775, 810, 850 and 910 nm) provided the light source. The diodes were pulsed in rapid succession and the light detected by the PMT. The use of four laser diodes enables more chromophores to be detected and also increases the sensitivity of the instrument (Mancini et al., 1994; Belardinelli et al., 1995). The intensity of incident and transmitted light was recorded continuously at 1 Hz and, along with the specific extinction coefficients and optical pathlength, used for on-line estimation and display of concentration changes (␮M) from the resting normoxic baseline of oxyhemoglobin ([HbO2 ]), deoxyhemoglobin ([HHb]) and total hemoglobin ([Hbtot ]) (Elwell, 1995). The values used for the differential pathlength factor (DPF) were 5.92 for cerebral (van der Zee et al., 1992) and 3.83 for m. vastus lateralis (Kowalchuk et al., 2002) monitoring. The HHb signal was assumed to be a reliable estimator of changes in tissue deoxygenation status (representing a mismatch between O2 delivery and O2 utilization) in the field of interrogation (De Blasi et al., 1994; Ferrari et al., 1997). Tissue oxygenation index (TOI (%) = HbO2 /Hbtot ) was calculated by NIRO 300 from the light attenuation slope along the distance from the emitting point as detected by the three sensors in the receiving optode. The raw attenuation signals (in optical density units) were transferred to a computer and stored for further analysis. 2.5. Local cerebral blood velocity (CBF velocity) Cerebral blood velocity was measured from the right MCA using transcranial Doppler ultrasound and a 2 MHz pulsed wave transducer (Multigon 500, Multigon Industries, Yonkers, NY) placed over the temporal window just above the zygomatic arch. Ultrasound gel was applied to the skin and hair of the area of the temporal window. Optimal Doppler signals from the MCA were achieved by varying the sample volume depth in incremental steps and at each step varying the angle of insonation to obtain the best quality signals. To ensure optimal insonation position and angle for the duration of the experiment the probe was securely positioned in a specially designed headband device. Instantaneous peak velocity frequency was recorded on a personal computer at 100 Hz together with ECG, oximetry, nearinfrared spectroscopy and mean arterial pressure (MAP) signals using a data acquisition and analysis system (Powerlab, ADInstruments, Colorado Springs, CO). During analysis, the R-peak of the ECG signal was used as a trigger to calculate mean of the peak velocities (Vmean ) for CBF on beat-by-beat basis. NIRS data, Vmean , fH, MAP and SpO2 %-values were averaged to give values in 1 s intervals and time-aligned with the gas exchange data. The mean values of the last 15 s of each test minute were chosen for statistical analysis. 2.6. Mean arterial pressure Arterial pressure was measured continuously from the finger of the left hand by photoplethysmographic methods (model 2300 Finapres, Ohmeda, Englewood, CA). The hand was maintained on the arm rest of the chair approximately 30 cm below the heart level throughout the testing period. Baseline blood

74

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

pressures from the Finapres device were corrected against manually obtained systolic and diastolic measures (brachial artery) before the onset of the data collection. Mean arterial pressure was calculated and used for further analysis. Although the diameter of the MCA would have been needed to calculate flow and cerebral vascular conductance, the change in cerebral vascular conductance was estimated as a ratio of Vmean and MAP (CVC = Vmean /MAP). CVC was estimated to determine if the vascular tree downstream from the middle cerebral artery was dilating or whether simple increases in blood pressure were causing the expected flow changes. 2.7. Indexes for chemosensitivity (AHVR), cerebrovascular responsiveness and tissue deoxygenation The individual chemosensitivity of ventilation was obtained by plotting ventilation against the SpO2 % on second-by-second basis and calculating a linear regression slope (l min−1 %−1 ) for that response (Harms and Stager, 1995; Guenette et al., 2004; Kolb et al., 2004b). The rate of change was also calculated for Vmean and TOI. The calculation of linear regression slope for these indexes was performed by using the mean values during the final 15 s of each minute. Thus, cerebrovascular responsiveness for progressive hypoxia was obtained by plotting Vmean versus SpO2 %, representing the increase in CBF velocity per unit decline in SpO2 % (cm s−1 %−1 ). Similarly, TOI versus SpO2 % represented the decline in TOIC and TOIM per unit decline in SpO2 % (% %−1 ). 2.8. Statistical analysis The effect of decreasing PETO2 and SpO2 % on measured variables was studied with repeated measures ANOVA. The values obtained during the AHVR test were compared with the initial normoxic values. When appropriate, Pearson correlation coefficient was used. Significance for all tests was established at P < 0.05, and data are expressed as means ± S.D. 3. Results 3.1. End-tidal O2 and CO2 , arterial O2 saturation, ventilation and hypoxic chemosensitivity The dynamic gas forcing system reduced PETO2 as expected while PETCO2 remained stable at the predetermined isocapnic level (Fig. 1). Arterial O2 saturation demonstrated an average reduction of 21 ± 6% during the AHVR test as SpO2 % decreased from 98 ± 1% in normoxia to 77 ± 7% at the end of the AHVR test (P < 0.001). In comparison to the initial normoxic level, the drop in SpO2 % was significant (P < 0.001) from the 5th minute onwards. At this point of the AHVR test, PETO2 was on average 73 ± 5 mmHg and SpO2 % of 94 ± 3%. As SpO2 % fell V˙ E and fH increased from their normoxic values of 12.9 ± 4.5 l min−1 and 70 ± 8 bpm to 25.2 ± 8.6 l min−1 and 82 ± 10 bpm, respectively. In comparison to normoxia, V˙ E was higher from the 7th minute (with mean SpO2 90 ± 4%) and fH from the 8th minute

Fig. 1. Cardiorespiratory responses during acute hypoxic ventilatory response test vs. arterial oxygen saturation. PETO2 () and PETCO2 () (top); V˙ E (䊉) (middle); Vmean () and HR () (bottom). Values are means ± S.D. Significantly different from normoxia  P < 0.05,  P < 0.01,  P < 0.001.

(with mean SpO2 87 ± 4%) onwards (Fig. 1). The mean value of hypoxic chemosensitivity was 0.52 ± 0.20 l min−1 %−1 as computed as an increase in V˙ E versus a percent decrease in SpO2 %. 3.2. Cerebral blood velocity, regional cerebral and muscle tissue oxygenation The initial normoxic value for Vmean was 53.4 ± 10.4 cm s−1 and it increased to 60.6 ± 11.6 cm s−1 at the end of the AHVR test (P < 0.001) (Fig. 1) with a mean slope of 0.33 ± 0.19 cm s−1 SpO2 %−1 . For HbtotC , the total increment was 2.97 ± 1.15 ␮M

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

75

In cerebral tissue, both HbO2 and TOI decreased and HHb increased as hypoxia became more severe. The total change from normoxia to the end-AHVR value for cerebral HHb was 5.55 ± 2.22 ␮M and for HbO2 −2.57 ± 1.99 ␮M, respectively. At the same time, TOIC decreased 6.2 ± 3.4% (all P < 0.001) (Fig. 2). Differences were found in the oxygenation profile of muscle tissue when compared to cerebral tissue. In contrast to the increase in HbtotC seen during the AHVR test, HbtotM did not change in hypoxia. Moreover, oxygenation was well protected in muscle despite reducing SpO2 % as HbO2M , HHbM and TOIM did not change significantly from the normoxic control level (Fig. 2). This difference between cerebral and muscle tissue oxygenation as a response to increasing hypoxia was also well described by the TOI/ SpO2 %-index (representing the decline in TOI per unit decline in SpO2 %), which was greater in cerebral (0.34 ± 0.14) than muscle tissue (0.07 ± 0.04) (P < 0.001). 3.3. Mean arterial pressure Blood pressure signal was recorded successfully from six of the eight subjects. In these subjects, MAP did not change from normoxia to the end of the hypoxic exposure, while the estimated CVC increased an average 11.5% (P < 0.05). 3.4. Relationships between variables The correlation coefficients between NIRS variables versus cerebrovascular responsiveness or hypoxic ventilatory chemosensitivity did not reach statistical significance. Similarly, the association between ventilatory chemosensitivity and cerebrovascular responsiveness failed to achieve statistical significance as the correlation coefficient between AHVR and Vmean slope was −0.64 (P = 0.087). 4. Discussion This study reports two main findings: first, cerebral tissue deoxygenation increased during the AHVR test, whereas the deoxygenation status of skeletal muscle remained at the initial normoxic level. Secondly, the degree of cerebral deoxygenation was not related to the cerebrovascular responsiveness or hypoxic ventilatory chemosensitivity, although both CBF velocity and ventilation increased as expected.

Fig. 2. Local cerebral (䊉) and muscle tissue () oxygenation profiles (HbO2 ; HHb; Hbtot and TOI%) during acute hypoxic ventilatory response test vs. arterial oxygen saturation. Values are means ± S.D. Significantly different from normoxia  P < 0.05,  P < 0.01,  P < 0.001.

with a mean increment of 0.13 ± 0.04 ␮M SpO2 %−1 (Fig. 2). On an individual level, both Vmean and HbtotC increased in each subject. However, the correlation coefficient of the slopes for the increase in HbtotC and Vmean failed to reach the level of statistical significance (r = 0.62, P = 0.09).

4.1. Regional cerebral blood velocity and the change in total Hb concentration During the AHVR test, Vmean increased in each subject and the increase was comparable with previously reported values during an acute hypoxic challenge (Jensen et al., 1996; Ainslie and Poulin, 2004; Kolb et al., 2004a,b). The fact that both Vmean and HbtotC increased during the AHVR test suggests increased flow in MCA, thus indicating the sensitivity of cerebral blood flow to acute isocapnic hypoxia. That the mutual correlation of Vmean and HbtotC failed to achieve statistical significance

76

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

(r = 0.62, P = 0.09) is comparable with the paper of Ide and Secher (2000) as both parameters also increase during exercise but the time scale and the relative changes are different. Moreover, the NIRS Hbtot signal is a volumetric signal and not a flow signal. Thus, it is possible to have a large increase in flow without a change in total Hb concentration as long as red cell spacing is not changed, there is no capillary recruitment or local vasodilation. An additional explanation for the non-significant correlation coefficient between Vmean and HbtotC is that the MCA also supplies other areas of the brain outside the area of NIRS interrogation and some of the area recorded by NIRS is supplied by the ACA. Thus, the change in Vmean in the middle cerebral artery is not necessarily accompanied by a similar change in HbtotC in that part of the frontal lobe where the NIRS signal was recorded. 4.2. Local cerebral and muscle oxygenation The human brain is a highly oxidative organ. At rest, the total O2 consumption for an average 1.4 kg human brain approximates 40–50 ml min−1 or 3.0–3.5 ml 100 g−1 min−1 (approximately 20% of bodily O2 consumption; Erecinska and Silver, 2001) while 30 kg of skeletal muscle consume approximately 60 ml O2 min−1 , or 0.2 ml 100 g−1 min−1 (Rowell, 1986). For cerebral grey matter, the rate of oxidative metabolism is even higher approximating 7 ml 100 g−1 min−1 (Levick, 2003). Previous animal studies have indicated that, during hypoxia, O2 consumption is unaffected (Weiss et al., 1983; Todd et al., 1994), arterial and venous O2 saturation decrease uniformly and to the same extent in the examined brain regions and that the cerebral O2 extraction rate is maintained (Weiss et al., 1983) or slightly but non-significantly increased (Todd et al., 1994). Thus, it appears that during hypoxemia, cerebral oxygenation is maintained primarily by increasing CBF with only minimal changes in O2 extraction rate. However, the issue is complicated as it has been suggested that there are differences in the O2 extraction between various brain areas and some areas, like hippocampus (having a key role in many aspects of cognitive function, including memory), may have less scope in which to increase extraction before having to increase flow (Dunn et al., 1999). The present study indicated major differences in the cerebral and muscle tissue deoxygenation status during acute progressive hypoxia at rest. During the AHVR test, the deoxygenation status of skeletal muscle was not changed whereas cerebral tissue showed changes reflecting an increased deoxygenation (i.e. increased HHb concentration). At the same time, TOIC reduced significantly from an average of 63.3% to 57.1% (i.e. 0.34% SpO2 %−1 ). This TOI/SpO2 %-index suggests that acute hypoxia may indeed affect cerebral tissue more than muscle tissue at rest as this index was almost five times greater in cerebral than muscle tissue. Both HHb and TOI responses indicate a local mismatch between O2 delivery and O2 utilization in cerebral tissue. The normoxic TOIC value is close to that (66 ± 8%) reported by Madsen and Secher (1999) but somewhat lower than in the recent study of Kolb et al. (2004b). Comparable to the trend seen in the present study, Kolb et al. observed a reduction from an average of 74% to 66% (0.80% SpO2 %−1 ) during an AHVR test.

The difference in the absolute %-values from different studies may well be due to differences in the measuring devices and slight differences in the areas of interrogation. The maintenance of local muscle oxygenation in acute hypoxia at rest is in accordance with the recent results from the same laboratory (DeLorey et al., 2004) but contrary to what was seen after an ascent to 3459 m as Imray et al. (2003) reported muscle saturation to fall from 73% at the control level (150 m) to 68% (P < 0.001) after a night at altitude. Their results of reduced and ours of maintained TOI suggest that the oxygenation response within muscle in hypoxia is time dependent so that in the very acute situation (i.e. minutes) oxygenation index is well maintained but decreased as hypoxia continues overnight (i.e. several hours). Moreover, the lack of changes in skeletal muscle deoxygenation most likely reflects a maintained balance between O2 delivery and O2 utilization. This finding may be explained by previous studies (e.g. Dinenno et al., 2003) showing that systemic hypoxia evokes significant limb vasodilatation, as indicated by increases in blood flow and vascular conductance. More precisely, while acute hypoxia causes sympathetic stimulation directly via stimulation of peripheral chemoreceptors and vagal withdrawal indirectly via increase in ventilation, local vascular beds seem to be regulated by arterial O2 content. The balance between the local effects of hypoxia and changes in the neural control of vascular tone determines the net vasodilatation or vasoconstriction in a vascular bed (Weisbrod et al., 2001). Thus, while heart rate and cardiac output increase in acute hypoxia and systemic vascular resistance decreases as all peripheral arteries except pulmonary arteries tend to vasodilate in acute hypoxia, their combined effect has been reported to decrease MAP transiently (Rowell and Blackmon, 1987; Wolfel and Levine, 2001), to maintain (Magosso and Ursino, 2004) or increase it (Halliwill and Minson, 2002). The observed nonsignificant increase (4%) in MAP was then well within the previous findings. In summary, despite a lower arterial O2 content in hypoxia, O2 delivery to the leg seems to be similar in hypoxia compared to normoxia (DeLorey et al., 2004). 4.3. Interaction between cerebral oxygenation, cerebrovascular responsiveness and ventilatory chemosensitivity The NIRS variables were not associated with cerebrovascular responsiveness or ventilatory chemosensitivity. To evaluate this finding, it should be emphasized that an efficient autoregulation adjusts cerebral blood flow in a wide range of physiological stimuli mainly by altering vascular resistance. Factors such as arterial partial pressures of CO2 and O2 can modulate resistance (Herholz et al., 1987; Giller et al., 1998; Ide and Secher, 2000; Vovk et al., 2002). The observed increase in estimated CVC (11.5%, P < 0.05) suggests that vasodilation contributed, in part, to the augmented Vmean . However, the diameter of the MCA changes very little, if any, in acute hypoxia (Poulin and Robbins, 1996; Kolb et al., 2004a). Thus, the augmented Vmean and CVC suggest that there is vasodilation occurring downstream in the arterioles that causes an increased velocity through the MCA (Hudetz, 1997).

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

The cerebrovascular responsiveness to hypoxia has been previously studied during a test of similar (Kolb et al., 2004b) and longer (Ellingsen et al., 1987) duration. We found the increase in Vmean as a response to reduced arterial saturation to be of similar magnitude as previously reported (Kolb et al., 2004b). However, a longer hypoxic exposure has revealed the hypoxic cerebrovascular responsiveness to be relatively slow, with an onresponse time constant (τ) about 6 min (Ellingsen et al., 1987). Therefore, it is evident that the whole response for Vmean was not achieved during the present study and the conclusion of this study should be limited to the present study conditions. In natural poikilocapnic high altitude, cerebral blood flow during the first 2–4 h appears to be slightly but not significantly increased (Huang et al., 1987). By 6–12 h a 24% (Severinghaus et al., 1966) and by 18–44 h a 20% increase (Huang et al., 1987) in CBF was observed, that was followed by a decline during the subsequent days to (Huang et al., 1987) or towards (Severinghaus et al., 1966) the values similar to those at sea level. Furthermore, cerebral blood flow and CBF velocity either remain relatively constant or increase slightly when PaO2 and SpO2 % are reduced moderately. Only when PaO2 is reduced more to a level of about 50–60 mmHg, would CBF increase substantially suggesting a threshold phenomenon for the increase in cerebral blood flow (Ainslie and Poulin, 2004; Magosso and Ursino, 2004). It has been also suggested that the hypoxia-induced increase in muscle blood flow occurs earlier at a higher PaO2 than the increase in cerebral blood flow (Magosso and Ursino, 2004). This may partly explain why muscle tissue maintained its oxygenation status better than cerebral tissue during the AHVR test. Moreover, those previous studies that have superimposed PaCO2 adjustment on acute hypoxia clearly indicate that the threshold for the hypoxia-induced increase in CBF occurs at higher PETO2 levels in hypoxic hypercapnia than during hypoxic iso- or poikilocapnia and that the greatest increase in CBF is seen during hypercapnia, moderate increase in isocapnia and a small increase in poikilocapnia (Ainslie and Poulin, 2004). When the relationship between Vmean and hypoxia is expressed as a function of SpO2 %, the relationship becomes relatively linear (Fig. 1) as previously noted (Kolb et al., 2004b). The increase in Vmean , however, was significant only at the end of the AHVR test. That Vmean responses during the present study were not related to cerebral deoxygenation should be interpreted to indicate that CBF is controlled most likely by a threshold mechanism under these conditions. This substantial increase in CBF only when PaO2 is reaching a critical level may be regarded as a sign that there would be enough O2 in the brain and a moderate acute reduction in oxygen delivery could be compensated by maintenance of oxygen extraction. Given that cerebral venous O2 saturation is about 60–70% at rest in normoxia (Henson et al., 1998; Kim et al., 2000; Watzman et al., 2000) and approximately 50% in hypoxia with PETO2 of 45 mmHg (Henson et al., 1998; Kim et al., 2000) this conclusion may be drawn. This conclusion seems especially tempting as the affinity of cytochrome c oxidase for O2 is very high, which under normal conditions ensures undiminished activity of oxidative phosphorylation down to very low PO2 . By contrast, many other enzymes have Km values for O2 within, or above, the ambient

77

cerebral gas tension, thus making their operations very dependent on O2 level in the physiological range (Erecinska and Silver, 2001) and this may be associated with the reduction in cognitive and motor function seen already at moderate altitude (see Section 4.4). As ventilation in isocapnic hypoxia is mainly regulated by peripheral chemoreceptors we did not expect strong causality between tissue oxygenation or deoxygenation and ventilation especially when PETO2 was adjusted and thus “blocking” the benefits of increased ventilation. In our study, the average increase in ventilation was 0.52 l min−1 SpO2 %−1 , which is in close agreement with the results in trained men (Guenette et al., 2004) but lower than in another study with healthy adult men (Kolb et al., 2004b). Although the correlation coefficient between the rate of increase in Vmean and AHVR failed to reach statistical significance (r = −0.64, P = 0.087), it suggests that the tendency for strong hypoxic ventilatory chemosensitivity is somewhat associated with blunted cerebrovascular responsiveness, i.e. the subjects with strong hypoxic ventilatory chemosensitivity may express smaller increase in cerebral blood flow even in isocapnic hypoxia than their less chemosensitive counterparts. Previously, Ainslie and Poulin (2004) found a positive correlation between hypoxic ventilatory chemosensitivity and cerebrovascular responsiveness in hypercapnic and isocapnic conditions while a negative correlation was seen in poikilocapnic hypoxia. Thus, despite well-maintained isocapnic hypoxia in their and our studies, a tendency for opposite relationship between ventilatory chemosensitivity and cerebrovascular responsiveness is evident. Therefore, further studies are needed to better understand the factors that are capable of modifying the relationship of ventilatory and cerebrovascular responsiveness. 4.4. Is the increase in cerebral deoxygenation meaningful? As this study indicates that cerebral deoxygenation is increased and cerebral tissue oxygenation index is decreased in acute hypoxia it is necessary to speculate on the functional relevance of this finding: is the deoxygenation meaningful or is there still enough reserve for normal brain function? In the present study, the changes from normoxia started to occur at the 5th minute with an average SpO2 % of 94 ± 3%, PETO2 73 ± 5 mmHg and inspired O2 tension (PIO2 ) 110 ± 2 mmHg. This PIO2 corresponds to an altitude of 3039 m, while the PIO2 of 67 ± 4 mmHg at the end of the AHVR test equates to an altitude of 6769 m. These hypoxic levels have a potential to alter brain function as previous studies clearly indicate slowed performance, particularly on more complex tests of cognitive and motor function, already at moderate altitude (Denison et al., 1966; Ernsting, 1978), although the changes are more prominent with increasing altitude (Hornbein, 2001). Moreover, the effect of progressive prolonged hypoxia during Operation Everest II indicated that cognitive disruption was more prevalent than motor (Kennedy et al., 1989). Further studies combining NIRS and TCD measurements with cognitive and motor tasks are warranted to help to expand understanding of the brain physiology.

78

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

4.5. Methodological considerations 4.5.1. AHVR test The ventilatory response to hypoxia varies not only between individuals but also within individuals (Sahn et al., 1977; Schoene, 1982) and the determined hypoxic ventilatory response is also affected by the method used to detect this phenomenon (Zhang and Robbins, 2000). Previously, a common way to express AHVR has been with a value A, which is a mathematical expression of the hyperbolic function derived from plotting V˙ E against PAO2 . With this model, Sahn et al. (1977) reported the within-day coefficient of variation (CV) to be on average 19%. Furthermore, the ventilatory response between days was more variable than within-day response and the changes in pH were expected to partially explain the between-day variance in AHVR (Sahn et al., 1977). One plausible reason to cause variation in the AHVR is the exponential fitting procedure to determine the value A. Therefore, using V˙ E versus SpO2 % instead of V˙ E versus PAO2 is a common way to express AHVR (Harms and Stager, 1995; Guenette et al., 2004; Kolb et al., 2004b) as it obviates the problem of how best to quantify the exponential hypoxic response (Weil and Zwillich, 1976). However, this linear fit model also produces variability within the AHVR and Zhang and Robbins (2000) reported a CV of 23% for AHVR in incremental step hypoxic exposure. Most recently, Kolb et al. (2004b) reported an average between-day CV of 15% for AHVR and a strong test–retest reproducibility (r = 0.93) value was reported also by Harms and Stager (1995). Nishimura et al. (1991) followed AHVR in 32 healthy male volunteers at intervals of 8–10 years and reported the response to be significantly correlated (r = 0.63, P < 0.001) between the initial and final examinations. Thus, it seems safe to conclude that although the AHVR contains variability within and between days the test–retest reproducibility is relatively good and the method therefore indicates individual differences in the hypoxic ventilatory chemosensitivity. 4.5.2. NIRS measurements Regional oxygenation of cerebral (frontal lobe) and muscle (m. vastus lateralis) tissue was recorded with NIRS. Previous data indicate that age, hemoglobin concentration, sensor location, equipment and the algorithm can affect cerebral saturation values (Owen-Reece et al., 1999; Kishi et al., 2003) and both good and poor reproducibility has been reported for the NIRS methodology. In the recent study of Kolb et al. (2004b), an average between-day CV for cerebral tissue saturation was 8%. This is in close agreement with the CV of 9.4% that Thavasothy et al. (2002) reported for absolute baseline values of cerebral oxygenation. Moreover, Koike et al. (2004) reported a good test–retest reproducibility for cerebral HbO2 during exercise (r = 0.88, P < 0.0001). In critically ill newborns and infants simple linear correlation analysis revealed significant associations between cerebral TOI and both invasively and transcutaneously measured arterial saturation, central venous saturation and arteriovenous O2 extraction (Weiss et al., 2005). In another study with neonates and infants, the cerebral TOI measurements after sensor exchange were not well reproducible under clinical conditions (Dullenkopf et al., 2005).

While the NIRS signal has been reported to correlate both roughly (Madsen and Secher, 1999) and closely (Mancini et al., 1994) to local venous O2 saturation, the contribution from capillaries and arterioles is significant and the NIRS signal provides for a close monitoring of tissue oxygenation (Madsen and Secher, 1999). The arterial and venous contribution to cerebral oximetry has been reported to average 16 ± 21% and 84 ± 21% with significant differences among subjects (Watzman et al., 2000). Likewise, for muscle at rest, the vascular compartments of the reflected blood volume are estimated to be proportioned 10:20:70% between arteriolar:capillary:venular segments, respectively (Boushel et al., 2001). However, since the NIRS technique requires light to penetrate the skin and subcutaneous fat, and in head also the skull, in order to reach the tissue of interest, changes in skin blood flow may alter the NIRS signal in a fashion unrelated to blood flow, O2 delivery and O2 utilization in the tissue of interest. Therefore, the contribution of extracranial oxygenation on regional cerebral saturation has been studied intensively. The evaluation of cerebral tissue saturation has been validated by comparing NIRS values with direct jugular vein blood gas measurements (SjvO2 ) during progressive hypoxia in humans. The overall method comparison analysis has produced a bias (mean value between the differences of the two methods) of −3.1% (Shah et al., 2000) and 5.2% (Kim et al., 2000) and a precision (S.D. of the differences) of 12.1% (Shah et al., 2000) and 10.7% (Kim et al., 2000), respectively. Thus, cerebral oxygen saturation measured by cerebral oximetry compares well to the measured SjvO2 in normal subjects, despite multiple physiological reasons for differences. These authors then concluded that regional cerebral saturation obtained by NIRS may serve as a reliable indicator of changes in brain oxygenation and deoxygenation induced by hypoxemia (Kim et al., 2000; Shah et al., 2000). Moreover, the closer relationship of SjvO2 to tissue than arterial saturation under the conditions of these experiments indicates that the measurement reflects primarily intracranial saturation (Kim et al., 2000). It has been also reported that during isocapnic hypoxia in healthy persons, cerebral oxygenation as estimated by NIRS precisely tracks changes in SjvO2 within individuals, but the relation exhibits a wide range of slopes and intercepts (Henson et al., 1998). Furthermore, Grubhofer et al. (1997) found that there was no correlation between regional O2 oxygenation and facial vein (draining substantially blood from forehead areas) oxygenation (P = 0.35) but there was a significant correlation between regional cerebral oxygenation and jugular venous bulb oxygenation (P = 0.027). Their linear regression analysis predicted a 3.6% change in regional cerebral oxygenation for every 10% change in jugular venous bulb oxygenation. They concluded that extracranial tissue oxygenation had a negligible influence on the values recorded using NIRS (Grubhofer et al., 1997). Also others (Samra et al., 1999) have hypothesized that the estimated concentration changes are dominated by changes of the oxygenation of the brain. A strong argument in favor of the applicability of NIRS measurements on intracranial oxygenation came from Owen-Reece et al. (1996). They used tourniquet inflation to occlude scalp blood flow and the occlusion was confirmed by laser Doppler velocimetry. Their results showed that tourniquet inflation had no effect on the

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

estimated value of cerebral blood volume or the differential pathlength factor. They concluded that, provided the distance between light entry and exit on the surface of the scalp is sufficiently large, changes in scalp blood flow have no effect on NIRS measurement of cerebral hemodynamics. On the contrary, some other studies (Klaessens et al., 2005) conclude that skin significantly influences regional cerebral oxygenation measurements. The extracranial versus intracranial contribution on NIRS values may be also affected by the equipments used and the positioning of the probes. The device used in the present study (NIRO 300) relies on a refinement of the Beer–Lambert law and incorporates a more rigorous application of spatially resolved spectroscopy (Matcher et al., 1995; Suzuki et al., 1999). Although it does not specifically attempt to exclude any contribution from extracranial blood, a degree of compensation does occur which is intrinsic to the algebraic differentiation of the attenuation gradient measured by the multichannel detector (Thavasothy et al., 2002). During carotid surgery, the sensitivity of cerebral TOI (NIRO 300) to intracranial changes has been reported to be 87.5% with a specificity of 100% (Al-Rawi et al., 2001). In contrast, the sensitivity and specificity to extracranial changes were 0%, suggesting that TOI measured by NIRO 300 reflects changes in cerebral tissue oxygenation (Al-Rawi et al., 2001; Thavasothy et al., 2002). The standard positioning of the probes is crucial to the success of NIRS in providing stable indexes of cerebral oxygenation change. The most practical approach (also applied in the present study) from the point of view of standardization would be to place the probes above and along the eyebrow. The distance between the optodes should be as high as 5–6 cm, and the probes should sample only from the MCA territory, avoiding sinuses and muscles (Smielewski et al., 1995). Similarly, skin blood flow represents a potential confounding factor when using NIRS to assess muscle tissue oxygenation and blood flow. Mancini et al. (1994) studied the contribution of skin flow to the changes in 760–800 nm absorption by simultaneous measurement of skin blood flow with laser flow Doppler and NIR recordings during hot water immersion. They found changes in skin flow but not in 760–800 nm absorption and concluded that NIRS measurements are minimally affected by skin blood flow. This opinion has been recently challenged (Minson, 2003; Lee and Clarke, 2004; Davis et al., 2006). Both local and whole body heating are reported to cause large increases in NIR-derived tissue oxygenation signal (Lee and Clarke, 2004; Davis et al., 2006), potentially confounding interpretation of the NIRS signal during conditions where both skin and muscle blood flows are elevated concomitantly (e.g. heat and/or prolonged exercise). The present study, however, was conducted at rest in normal room temperature with obviously minimal thermal effects on skin blood flow. Nevertheless, hypoxia itself may affect skin blood flow as in thermoneutral conditions non-acral skin appears to vasodilate in hypoxia, whereas acral skin vasoconstricts (Minson, 2003). In conclusion, the usability of NIRS methodology is dependent on several factors like the equipment itself, the optode positioning and the inter-optode distance and the algorithm used for calculation of the results. It seems safe to conclude that, although NIRS methodology produces variation within the results, at least

79

tracking trends in cerebral and muscle oxygenation would be acceptable (Henson et al., 1998). 4.5.3. TCD measurements Transcranial Doppler (TCD) was used as a non-invasive method to study cerebral blood flow. Although TCD does not provide a direct measure of CBF, the changes in Vmean reflect changes in CBF, provided that the diameter of the insonated artery remains constant (Bishop et al., 1986). In fact, previous studies (Poulin and Robbins, 1996; Kolb et al., 2004a) suggest that there is little, if any, change in the cross-sectional area of the middle cerebral artery during hypoxic exposure. The validation of TCD is based on the finding that TCD measured changes in CBF velocity reliably correlate with changes in CBF (133 Xenon technique) but the absolute velocity cannot be used as an indicator of CBF (Bishop et al., 1986). For CBF velocity measurements with TCD from the MCA, the average CV for test–retest Vmean was reported to be 10.4% (Kolb et al., 2004b). Similarly, in the interobserver study in which CBF velocity measurements were repeated within hours, r was 0.92 and CV was 8.8%, and in the intraobserver study in which measurements were repeated with a 2-month interval, r was 0.80 and CV was 13.0% (Bay-Hansen et al., 1997). 4.6. Summary In the current model, the ventilatory response to acute hypoxia was not allowed to adjust arterial Hb saturation. Therefore, the observations made here reflect cardiovascular adjustment in cerebral and muscle circulation during the AHVR test. Despite an increase in regional CBF velocity, local total hemoglobin concentration, heart rate and ventilation it was observed that cerebral tissue reflected signs of deoxygenation during the acute isocapnic hypoxic ventilatory response test. However, the degree of cerebral deoxygenation was not related to the cerebrovascular responsiveness or ventilatory chemosensitivity. At the same time, muscle tissue deoxygenation status was maintained at the initial normoxic levels. Therefore, tissuespecific control features appear to exist that favor muscle perfusion during this test. Acknowledgements We thank our subjects for their enthusiastic participation. The technical help of Brad Hansen is highly acknowledged. This study was supported by the National Science and Engineering Research Council (NSERC, Canada), The Heart and Stroke Foundation of Ontario (Grant #T5342), Ministry of Education (Finland), and Finnish Cultural Foundation. References Ainslie, P.N., Poulin, M.J., 2004. Ventilatory, cerebrovascular, and cardiovascular interactions in acute hypoxia: regulation by carbon dioxide. J. Appl. Physiol. 97, 149–159. Al-Rawi, P.G., Smielewski, P., Kirkpatrick, P.J., 2001. Evaluation of a nearinfrared spectrometer (NIRO 300) for the detection of intracranial oxygenation changes in the adult head. Stroke 32, 2492–2500.

80

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81

Bay-Hansen, J., Ravn, T., Knudsen, G.M., 1997. Application of interhemispheric index for transcranial Doppler sonography velocity measurements and evaluation of recording time. Stroke 28, 1009–1014. Beaver, W.L., Lamarra, N., Wasserman, K., 1981. Breath-by-breath measurement of true alveolar gas exchange. J. Appl. Physiol. 51, 1662–1675. Belardinelli, R., Barstow, T.J., Porszasz, J., Wasserman, K., 1995. Skeletal muscle oxygenation during constant work rate exercise. Med. Sci. Sports Exerc. 27, 512–519. Bishop, C.C., Powell, S., Rutt, D., Browse, N.L., 1986. Transcranial Doppler measurement of middle cerebral artery blood flow velocity: a validation study. Stroke 17, 913–915. Boushel, R., Langberg, H., Olesen, J., Gonzales-Alonzo, J., Bulow, J., Kjaer, M., 2001. Monitoring tissue oxygen availability with near infrared spectroscopy (NIRS) in health and disease. Scand. J. Med. Sci. Sports 11, 213–222. Burton, M.D., Kazemi, H., 2000. Neurotransmitters in central respiratory control. Respir. Physiol. 122, 111–121. Cohen, P.J., Alexander, S.C., Smith, T.C., Reivich, M., Wollman, H., 1967. Effects of hypoxia and normocarbia on cerebral blood flow and metabolism in conscious man. J. Appl. Physiol. 23, 183–189. Davis, S.L., Fadel, P.J., Cui, J., Thomas, G.D., Crandall, C.G., 2006. Skin blood flow influences near-infrared spectroscopy-derived measurements of tissue oxygenation during heat stress. J. Appl. Physiol. 100, 221–224. De Blasi, R.A., Ferrari, M., Natali, A., Conti, G., Mega, A., Gasparetto, A., 1994. Noninvasive measurement of forearm blood flow and oxygen consumption by near-infrared spectroscopy. J. Appl. Physiol. 76, 1388–1393. DeLorey, D.S., Shaw, C.N., Shoemaker, J.K., Kowalchuk, J.M., Paterson, D.H., 2004. The effect of hypoxia on pulmonary O2 uptake, leg blood flow and muscle deoxygenation during single-leg knee-extension exercise. Exp. Physiol. 89, 293–302. Denison, D.M., Ledwith, F., Poulton, E.C., 1966. Complex reaction times at simulated cabin altitudes of 5,000 feet and 8,000 feet. Aerosp. Med. 37, 1010–1013. Dinenno, F.A., Joyner, M.J., Halliwill, J.R., 2003. Failure of systemic hypoxia to blunt alpha-adrenergic vasoconstriction in the human forearm. J. Physiol. 549, 985–994. Dullenkopf, A., Kolarova, A., Schulz, G., Frey, B., Baenziger, O., Weiss, M., 2005. Reproducibility of cerebral oxygenation measurement in neonates and infants in the clinical setting using the NIRO 300 oximeter. Pediatr. Crit. Care Med. 6, 344–347. Dunn, J.F., Wadghiri, Y.Z., Meyerand, M.E., 1999. Regional heterogeneity in the brain’s response to hypoxia measured using BOLD MR imaging. Magn. Reson. Med. 41, 850–854. Ellingsen, I., Hauge, A., Nicolaysen, G., Thoresen, M., Walloe, L., 1987. Changes in human cerebral blood flow due to step changes in PAO2 and PACO2 . Acta Physiol. Scand. 129, 157–163. Elwell, C., 1995. A Practical Users Guide to Near Infrared Spectroscopy. Hamamatsu Photonics KK, London, pp. 1–155. Erecinska, M., Silver, I.A., 2001. Tissue oxygen tension and brain sensitivity to hypoxia. Respir. Physiol. 128, 263–276. Ernsting, J., 1978. Prevention of hypoxia—acceptable compromises. Aviat. Space Environ. Med. 49, 495–502. Ferrari, M., Binzoni, T., Quaresima, V., 1997. Oxidative metabolism in muscle. Philos. Trans. R. Soc. Lond. B Biol. Sci. 352, 677–683. Gavin, T.P., Derchak, P.A., Stager, J.M., 1998. Ventilation’s role in the decline in VO2max and SaO2 in acute hypoxic exercise. Med. Sci. Sports Exerc. 30, 195–199. Gibson, G.E., Pulsinelli, W., Blass, J.P., Duffy, T.E., 1981. Brain dysfunction in mild to moderate hypoxia. Am. J. Med. 70, 1247–1254. Giller, C.A., Hatab, M.R., Giller, A.M., 1998. Estimation of vessel flow and diameter during cerebral vasospasm using transcranial Doppler indices. Neurosurgery 42, 1076–1081. Grubhofer, G., Lassnigg, A., Manlik, F., Marx, E., Trubel, W., Hiesmayr, M., 1997. The contribution of extracranial blood oxygenation on nearinfrared spectroscopy during carotid thrombendarterectomy. Anaesthesia 52, 116–120. Guenette, J.A., Diep, T.T., Koehle, M.S., Foster, G.E., Richards, J.C., Sheel, A.W., 2004. Acute hypoxic ventilatory response and exercise-induced arterial hypoxemia in men and women. Respir. Physiol. Neurobiol. 143, 37–48.

Halliwill, J.R., Minson, C.T., 2002. Effect of hypoxia on arterial baroreflex control of heart rate and muscle sympathetic nerve activity in humans. J. Appl. Physiol. 93, 857–864. Harms, C.A., Stager, J.M., 1995. Low chemoresponsiveness and inadequate hyperventilation contribute to exercise-induced hypoxemia. J. Appl. Physiol. 79, 575–580. Henson, L.C., Calalang, C., Temp, J.A., Ward, D.S., 1998. Accuracy of a cerebral oximeter in healthy volunteers under conditions of isocapnic hypoxia. Anesthesiology 88, 58–65. Herholz, K., Buskies, W., Rist, M., Pawlik, G., Hollmann, W., Heiss, W.D., 1987. Regional cerebral blood flow in man at rest and during exercise. J. Neurol. 234, 9–13. Hornbein, T.F., 2001. The high-altitude brain. J. Exp. Biol. 204, 3129–3132. Hornbein, T.F., Townes, B.D., Schoene, R.B., Sutton, J.R., Houston, C.S., 1989. The cost to the central nervous system of climbing to extremely high altitude. N. Engl. J. Med. 321, 1714–1719. Huang, S.Y., Moore, L.G., McCullough, R.E., McCullough, R.G., Micco, A.J., Fulco, C., Cymerman, A., Manco-Johnson, M., Weil, J.V., Reeves, J.T., 1987. Internal carotid and vertebral arterial flow velocity in men at high altitude. J. Appl. Physiol. 63, 395–400. Hudetz, A.G., 1997. Blood flow in the cerebral capillary network: a review emphasizing observations with intravital microscopy. Microcirculation 4, 233–252. Ide, K., Secher, N.H., 2000. Cerebral blood flow and metabolism during exercise. Prog. Neurobiol. 61, 397–414. Imray, C.H., Walsh, S., Clarke, T., Tiivas, C., Hoar, H., Harvey, T.C., Chan, C.W., Forster, P.J., Bradwell, A.R., Wright, A.D., Society, B.M.R.E., 2003. Effects of breathing air containing 3% carbon dioxide, 35% oxygen or a mixture of 3% carbon dioxide/35% oxygen on cerebral and peripheral oxygenation at 150 m and 3459 m. Clin. Sci. (Lond.) 104, 203–210. Jensen, J.B., Sperling, B., Severinghaus, J.W., Lassen, N.A., 1996. Augmented hypoxic cerebral vasodilation in men during 5 days at 3,810 m altitude. J. Appl. Physiol. 80, 1214–1218. Katayama, K., Sato, Y., Morotome, Y., Shima, N., Ishida, K., Mori, S., Miyamura, M., 2001. Intermittent hypoxia increases ventilation and SaO2 during hypoxic exercise and hypoxic chemosensitivity. J. Appl. Physiol. 90, 1431–1440. Kennedy, R.S., Dunlap, W.P., Banderet, L.E., Smith, M.G., Houston, C.S., 1989. Cognitive performance deficits in a simulated climb of Mount Everest: Operation Everest II. Aviat. Space Environ. Med. 60, 99– 104. Kim, M.B., Ward, D.S., Cartwright, C.R., Kolano, J., Chlebowski, S., Henson, L.C., 2000. Estimation of jugular venous O2 saturation from cerebral oximetry or arterial O2 saturation during isocapnic hypoxia. J. Clin. Monit. Comput. 16, 191–199. Kishi, K., Kawaguchi, M., Yoshitani, K., Nagahata, T., Furuya, H., 2003. Influence of patient variables and sensor location on regional cerebral oxygen saturation measured by INVOS 4100 near-infrared spectrophotometers. J. Neurosurg. Anesthesiol. 15, 302–306. Klaessens, J.H., Hopman, J.C., Liem, K.D., van Os, S.H., Thijssen, J.M., 2005. Effects of skin on bias and reproducibility of near-infrared spectroscopy measurement of cerebral oxygenation changes in porcine brain. J. Biomed. Opt. 10, 44003. Koike, A., Itoh, H., Oohara, R., Hoshimoto, M., Tajima, A., Aizawa, T., Fu, L.T., 2004. Cerebral oxygenation during exercise in cardiac patients. Chest 125, 182–190. Kolb, J.C., Ainslie, P.N., Ide, K., Poulin, M.J., 2004a. Effects of five consecutive nocturnal hypoxic exposures on the cerebrovascular responses to acute hypoxia and hypercapnia in humans. J. Appl. Physiol. 96, 1745–1754. Kolb, J.C., Ainslie, P.N., Ide, K., Poulin, M.J., 2004b. Protocol to measure acute cerebrovascular and ventilatory responses to isocapnic hypoxia in humans. Respir. Physiol. Neurobiol. 141, 191–199. Kowalchuk, J.M., Rossiter, H.B., Ward, S.A., Whipp, B.J., 2002. The effect of resistive breathing on leg muscle oxygenation using near-infrared spectroscopy during exercise in men. Exp. Physiol. 87, 601–611. Lee, S., Clarke, M., 2004. Elevated skin blood flow influences near infrared spectroscopy measurement during supine rest. NASA Technical Paper 213149, S-942.

J.E. Peltonen et al. / Respiratory Physiology & Neurobiology 155 (2007) 71–81 Levick, J.R., 2003. An Introduction To Cardiovascular Physiology. Arnold, London. Madsen, P.L., Secher, N.H., 1999. Near-infrared oximetry of the brain. Prog. Neurobiol. 58, 541–560. Magosso, E., Ursino, M., 2004. Modelling study of the acute cardiovascular response to hypocapnic hypoxia in healthy and anaemic subjects. Med. Biol. Eng. Comput. 42, 158–166. Mancini, D.M., Bolinger, L., Li, H., Kendrick, K., Chance, B., Wilson, J.R., 1994. Validation of near-infrared spectroscopy in humans. J. Appl. Physiol. 77, 2740–2747. Matcher, S., Kirkpatrick, P., Nahid, K., Cope, M., Delpy, D., 1995. Absolute quantification methods in tissue near-infrared spectroscopy. In: Proceedings SPIE, 2389, pp. 486–495. Minson, C.T., 2003. Hypoxic regulation of blood flow in humans. Skin blood flow and temperature regulation. Adv. Exp. Med. Biol. 543, 249–262. Nishimura, M., Yamamoto, M., Yoshioka, A., Akiyama, Y., Kishi, F., Kawakami, Y., 1991. Longitudinal analyses of respiratory chemosensitivity in normal subjects. Am. Rev. Respir. Dis. 143, 1278–1281. Owen-Reece, H., Elwell, C.E., Wyatt, J.S., Delpy, D.T., 1996. The effect of scalp ischaemia on measurement of cerebral blood volume by near-infrared spectroscopy. Physiol. Meas. 17, 279–286. Owen-Reece, H., Smith, M., Elwell, C.E., Goldstone, J.C., 1999. Near infrared spectroscopy. Br. J. Anaesth. 82, 418–426. Poulin, M.J., Cunningham, D.A., Paterson, D.H., Kowalchuk, J.M., Smith, W.D., 1993. Ventilatory sensitivity to CO2 in hyperoxia and hypoxia in older aged humans. J. Appl. Physiol. 75, 2209–2216. Poulin, M.J., Robbins, P.A., 1996. Indexes of flow and cross-sectional area of the middle cerebral artery using doppler ultrasound during hypoxia and hypercapnia in humans. Stroke 27, 2244–2250. Raichle, M.E., Hornbein, T.F., 2001. The high-altitude brain. In: Hornbein, T.F., Schoene, R.B. (Eds.), High Altitude: An Exploration of Human Adaptation. Marcel Dekker Inc., New York, pp. 377–423. Roach, R.C., Hackett, P.H., 2001. Frontiers of hypoxia research: acute mountain sickness. J. Exp. Biol. 204, 3161–3170. Rolett, E.L., Azzawi, A., Liu, K.J., Yongbi, M.N., Swartz, H.M., Dunn, J.F., 2000. Critical oxygen tension in rat brain: a combined (31)P-NMR and EPR oximetry study. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279, R9–R16. Rowell, L.B., 1986. Human Circulation Regulation During Physical Stress. Oxford University Press, New York. Rowell, L.B., Blackmon, J.R., 1987. Human cardiovascular adjustments to acute hypoxaemia. Clin. Physiol. 7, 349–376. Sahn, S.A., Zwillich, C.W., Dick, N., McCullough, R.E., Lakshminarayan, S., Weil, J.V., 1977. Variability of ventilatory responses to hypoxia and hypercapnia. J. Appl. Physiol. 43, 1019–1025. Samra, S.K., Stanley, J.C., Zelenock, G.B., Dorje, P., 1999. An assessment of contributions made by extracranial tissues during cerebral oximetry. J. Neurosurg. Anesthesiol. 11, 1–5. Schoene, R.B., 1982. Control of ventilation in climbers to extreme altitude. J. Appl. Physiol. 53, 886–890. Severinghaus, J.W., Chiodi, H., Eger 2nd, E.I., Brandstater, B., Hornbein, T.F., 1966. Cerebral blood flow in man at high altitude. Role of cerebrospinal fluid pH in normalization of flow in chronic hypocapnia. Circ. Res. 19, 274–282.

81

Shah, N., Trivedi, N.K., Clack, S.L., Shah, M., Shah, P.P., Barker, S., 2000. Impact of hypoxemia on the performance of cerebral oximeter in volunteer subjects. J. Neurosurg. Anesthesiol. 12, 201–209. Smielewski, P., Kirkpatrick, P., Minhas, P., Pickard, J.D., Czosnyka, M., 1995. Can cerebrovascular reactivity be measured with near-infrared spectroscopy? Stroke 26, 2285–2292. Smith, C., Dempsey, J., Hornbein, T., 2001. Control of breathing at high altitude. In: Hornbein, T., Schoene, R.B. (Eds.), High Altitude: An Exploration of Human Adaptation. Marcel Dekker Inc., New York, pp. 139–173. Suzuki, S., Takasaki, S., Ozaki, T., Kobayashi, Y., 1999. Tissue oxygenation monitor using NIR spatially resolved spectroscopy. In: Chance, B., Alfano, R.R., Tromberg, B.J. (Eds.), Optical Tomography and Spectroscopy of Tissue III, vol. 3597. SPIE, pp. 582–592. Thavasothy, M., Broadhead, M., Elwell, C., Peters, M., Smith, M., 2002. A comparison of cerebral oxygenation as measured by the NIRO 300 and the INVOS 5100 near-infrared spectrophotometers. Anaesthesia 57, 999– 1006. Todd, M.M., Wu, B., Maktabi, M., Hindman, B.J., Warner, D.S., 1994. Cerebral blood flow and oxygen delivery during hypoxemia and hemodilution: role of arterial oxygen content. Am. J. Physiol. 267, H2025–H2031. van der Zee, P., Cope, M., Arridge, S.R., Essenpreis, M., Potter, L.A., Edwards, A.D., Wyatt, J.S., McCormick, D.C., Roth, S.C., Reynolds, E.O., 1992. Experimentally measured optical pathlengths for the adult head, calf and forearm and the head of the newborn infant as a function of inter optode spacing. Adv. Exp. Med. Biol. 316, 143–153. Vovk, A., Cunningham, D.A., Kowalchuk, J.M., Paterson, D.H., Duffin, J., 2002. Cerebral blood flow responses to changes in oxygen and carbon dioxide in humans. Can. J. Physiol. Pharmacol. 80, 819–827. Watzman, H.M., Kurth, C.D., Montenegro, L.M., Rome, J., Steven, J.M., Nicolson, S.C., 2000. Arterial and venous contributions to near-infrared cerebral oximetry. Anesthesiology 93, 947–953. Weil, J.V., Byrne-Quinn, E., Sodal, I.E., Friesen, W.O., Underhill, B., Filley, G.F., Grover, R.F., 1970. Hypoxic ventilatory drive in normal man. J. Clin. Invest. 49, 1061–1072. Weil, J.V., Zwillich, C.W., 1976. Assessment of ventilatory response to hypoxia: methods and interpretation. Chest 70, 124–128. Weisbrod, C.J., Minson, C.T., Joyner, M.J., Halliwill, J.R., 2001. Effects of regional phentolamine on hypoxic vasodilatation in healthy humans. J. Physiol. 537, 613–621. Weiss, H.R., Buchweitz, E., Sinha, A.K., 1983. Effect of hypoxic-hypocapnia on cerebral regional oxygen consumption and supply. Microvasc. Res. 25, 194–204. Weiss, M., Dullenkopf, A., Kolarova, A., Schulz, G., Frey, B., Baenziger, O., 2005. Near-infrared spectroscopic cerebral oxygenation reading in neonates and infants is associated with central venous oxygen saturation. Paediatr. Anaesth. 15, 102–109. Wolfel, E., Levine, B.D., 2001. The cardiovascular system at high altitude. In: Hornbein, T., Schoene, R. (Eds.), High Altitude: An Exploration of Human Adaptation. Marcel Dekker Inc., New York, pp. 235–292. Zhang, S., Robbins, P.A., 2000. Methodological and physiological variability within the ventilatory response to hypoxia in humans. J. Appl. Physiol. 88, 1924–1932.