Journal of Applied Physiology Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 100: 850-857, 2006. First published November 17, 2005; doi:10.1152/japplphysiol.00830.2005
8750-7587/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
100/3/850    most recent
00830.2005v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tichauer, K. M.
Right arrow Articles by Lawrence, K. St.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tichauer, K. M.
Right arrow Articles by Lawrence, K. St.

Near-infrared spectroscopy measurements of cerebral blood flow and oxygen consumption following hypoxia-ischemia in newborn piglets

Kenneth M. Tichauer,1,3 Derek W. Brown,1,2,3 Jennifer Hadway,1,2 Ting-Yim Lee,1,2,3 and Keith St. Lawrence1,2,3

1Imaging Division, Lawson Health Research Institute, London, Ontario; 2Imaging Research Laboratories, Robarts Research Institute, London, Ontario; 3Department of Medical Biophysics, University of Western Ontario, London, Ontario, Canada

Submitted 12 July 2005 ; accepted in final form 17 November 2005


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Impaired oxidative metabolism following hypoxia-ischemia (HI) is believed to be an early indicator of delayed brain injury. The cerebral metabolic rate of oxygen (CMRO2) can be measured by combining near-infrared spectroscopy (NIRS) measurements of cerebral blood flow (CBF) and cerebral deoxy-hemoglobin concentration. The ability of NIRS to measure changes in CMRO2 following HI was investigated in newborn piglets. Nine piglets were subjected to 30 min of HI by occluding both carotid arteries and reducing the fraction of inspired oxygen to 8%. An additional nine piglets served as sham-operated controls. Measurements of CBF, oxygen extraction fraction (OEF), and CMRO2 were obtained at baseline and at 6 h after the HI insult. Of the three parameters, only CMRO2 showed a persistent and significant change after HI. Five minutes after reoxygenation, there was a 28 ± 12% (mean ± SE) decrease in CMRO2, a 72 ± 50% increase in CBF, and a 56 ± 19% decrease in OEF compared with baseline (P < 0.05). By 30 min postinsult and for the remainder of the study, there were no significant differences in CBF and OEF between control and insult groups, whereas CMRO2 remained depressed throughout the 6-h postinsult period. This study demonstrates that NIRS can measure decreases in CMRO2 caused by HI. The results highlight the potential for NIRS to be used in the neonatal intensive care unit to detect delayed brain damage.

deoxyhemoglobin; cerebral blood volume; indocyanine green; oxygen extraction fraction

HYPOXIA-ISCHEMIA (HI) affects roughly 1 in every 250 births and continues to be a major cause of perinatal brain injury (35). The long-term effects of this insult include mental retardation, seizure disorders, and cerebral palsy (21). It is believed that a significant fraction of cell death induced by HI occurs 6–24 h after the initial injury (1). The delayed injury is attributed to apoptosis and has led to the concept of a "therapeutic window" (10). That is, there is a brief period after HI when it may be possible to prevent the cascade of molecular events that lead to apoptotic cell death and thereby minimize further brain injury. A considerable effort has been focused on finding methods to prevent the onset of apoptosis, with a number of treatments showing promising results (13, 24). However, considering the brevity of the therapeutic window, the efficacy of such treatments will require early detection (10).

A change in cerebral energy metabolism is one possible predictor of delayed brain injury (14). Prolonged HI impairs mitochondrial function, triggering the production of apoptotic initiators such as caspases and cytochrome c (14). Consequently, impaired energy metabolism after HI indicates the onset of delayed brain injury (2, 30). Clinical studies involving magnetic resonance spectroscopy (MRS), with both hydrogen (1H) and phosphorus (31P), have demonstrated changes in energy metabolism in newborns suspected of HI injury (6, 36); this has also been observed in animal models of HI (18, 26). Other indications of impaired energy metabolism include a marked increase in anaerobic glucose consumption, reduced oxygen consumption, impaired mitochondrial respiration, and reduced cytochrome oxidase activity (8, 11, 12, 29).

Only MRS offers a clinically feasible method of detecting changes in energy metabolism. However, MRS suffers from practical issues such as long examination periods and involves transportation of acutely ill infants who often require ventilation, multiple in-dwelling catheters, infusion, and vasopressor support. Because of these issues, the number of measurements that can be obtained during the critical first few hours of life is limited.

Near-infrared spectroscopy (NIRS) offers a means of measuring brain function at the bedside, due to its portability and because of recent advances in hemodynamic and metabolic measurements. With broadband spectrum NIRS, absolute concentrations of deoxyhemoglobin ([Hb]) can be measured using the second derivative approach (9). Cerebral blood flow (CBF) can be determined using indocyanine green (ICG) as an intravascular tracer (4, 23, 33). Our group has developed and validated a NIRS method for measuring the cerebral metabolic rate of oxygen (CMRO2) by combining NIRS measurements of CBF and [Hb] (5, 34). The purpose of the present study was to investigate the use of NIRS for detecting reductions in CMRO2 after HI. Experiments were conducted in a newborn animal model (piglets), and the postinsult CMRO2 values measured with NIRS were validated by independently measuring CMRO2 directly from blood samples.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Animal preparation.   The study was approved by the Council on Animal Care at the University of Western Ontario. Newborn Duroc piglets (n = 18) were delivered from a local supplier on the morning of the experiment. Piglets were anesthetized with isoflurane (3–4% during preparatory surgery, 1.5% postsurgery), paralyzed via injections of vecuronium (0.1 mg/kg every 30 min), tracheotomized, and mechanically ventilated on an oxygen-medical air mixture. Two incisions were made lateral to the trachea, and vascular occluders with an inner diameter of 1.5 mm were placed around the carotid arteries posterior to the clavicle (In Vivo Metric, Healdsburg, CA). Catheters were inserted into an ear vein for ICG injection and into a femoral artery to continuously monitor blood pressure and for the collection of arterial blood samples for gas and glucose analyses. All incisions were treated with marcaine.

Arterial CO2 tension was monitored throughout the experiment, either directly from blood samples or indirectly from the end-tidal CO2 tension, and maintained at ~40 Torr by adjusting the breathing rate. Arterial oxygen tension was maintained at a level of 100–130 Torr by adjusting the ratio of oxygen to medical air. A water heating blanket was used to maintain rectal temperature between 37.5 and 38.5°C. Blood glucose levels were monitored intermittently; if they fell below 3 mM, a 2-ml infusion of 25% glucose solution was administered intravenously. Other physiological parameters monitored throughout the experiment included arterial pH and heart rate. After surgery, piglets were positioned into a frame holding the NIRS probes and allowed to stabilize for 1 h before the beginning of the experiment.

Experimental procedure.   Piglets were randomly divided into two groups: the HI group (9 piglets, mean age 7.0 ± 3.5 h and median weight 1.68 ± 0.21 kg) and the control group (9 piglets, mean age 7.2 ± 2.6 h and median weight 1.52 ± 0.28 kg). After the stabilization period, piglets in the HI group were exposed to 30 min of HI by clamping both carotid arteries and lowering the inspired concentration of O2 (FIO2) to 8% (8). At the end of the HI insult, the carotid clamps were released and FIO2 was returned to baseline. For the control group, FIO2 was maintained at baseline levels and carotids were left unclamped, but all other procedures were kept identical to those of the HI group.

CBF, cerebral blood volume (CBV), total blood-hemoglobin concentration, arterial oxygen saturation (SaO2), cerebral [Hb], oxygen extraction fraction (OEF), and CMRO2 were measured periodically throughout the experiment. The measurements were collected three times at baseline, three times during the HI insult, at 5 min after the insult, and periodically for a further 6 h postinsult. During this period, measurements were collected at intervals of 30 min for the first 2 h and at intervals of 60 min for the remaining 4 h. The three measurements collected during HI were acquired at 5, 15, and 25 min within the insult period. A time interval of 10 min is not sufficient for complete ICG clearance between measurements. However, Eq. 1 remains valid because the residue ICG concentration can be considered to be at a quasi-steady-state level during the brief measuring intervals (4, 23). At the end of the experiment, the piglets were euthanized with an intravenous injection of potassium chloride.

Accuracy of CMRO2 measurement after HI.   An additional six piglets (mean age 16 ± 4 h and mean weight 1.7 ± 0.1 kg) were used to evaluate the accuracy of the NIRS CMRO2 measurements after the 30-min insult. In each piglet, the sagittal and lambdoid sutures were identified and a small burr hole was drilled 1 cm posterior to the lambda structure. A catheter was then inserted into the superior sagittal sinus for the collection of cerebral venous blood. All other surgical procedures, the time course of the experiment, and the duration of the HI insult were the same as described above.

The arterial-venous oxygen difference was determined from blood samples collected from a femoral artery and the sagittal sinus. This measurement was combined with the NIRS measurements of CBF to calculate a blood sample-based measurement of CMRO2 (Eq. 3). Blood sample- and NIRS-based measurements of CMRO2 were acquired concurrently and averaged over three repeated measurements at baseline, 3 h after HI, and 5 h after HI.

NIRS.   The NIRS system was a continuous wave, broadband (600–980 nm) unit composed of a tungsten halogen light source, two fiber-optic cables, and a spectrometer. The spectrometer consisted of a holographic grating housed in a light-tight container attached to a cooled charge-coupled device camera (Wright Instruments, Enfield, Middlesex, UK). To obtain measurements of CBF, CBV, and CMRO2, the two fiber-optic cables were placed 3.0 cm apart, biparietally on the head of the piglet near the coronal sutures. Near-infrared light was transmitted through one cable, and a fraction of the scattered light was collected by the second cable. The intensity spectrum of the collected light was determined by channeling the light onto the holographic grating where it was dispersed across the charge-coupled device chip with a spectral sampling width of 0.395 ± 0.001 nm. The concentrations of ICG and Hb, acquired with a temporal resolution of 200 ms, were based on a modified version of the Beer-Lambert law:

Formula 1(1)
where A({lambda}) is the absorbance of NIR light and {epsilon}i({lambda}) is the extinction coefficient of the ith chromophore at wavelength {lambda}, ci is the concentration of the ith chromophore, B is the distance the NIR light traveled (which is greater than the distance between the probes due to scatter), and G represents the amount of light lost to scatter. The differential path length and the absolute concentration of Hb were determined by using the second derivative technique described by Matcher et al. (19) to account for scatter and assuming a water concentration of 85% (9). Absolute concentrations of other NIR absorbers such as oxyhemoglobin and cytochrome oxidase cannot be confidently determined with the second derivative approach (9). The absolute change in the concentration of ICG was calculated using

Formula 2(2)
where {Delta}c is the change in the concentration of ICG, {Delta}A is the change in attenuation of the spectra, and {epsilon}ICG({lambda}) is the extinction coefficient spectrum for ICG.

Cerebral hemodynamics.   For a complete discussion on the use of NIRS for measuring CBF and CBV, see Brown et al. (4). Briefly, each measurement required a 0.1 mg/kg intravenous injection of the NIR chromophore ICG (23). The relationship between the accumulation of tracer in tissue, Q(t), and the concentration in arterial blood, Ca(t), is given by the following expression (37)

Formula 3(3)
where * is the convolution operator. Q(t) was determined by the NIRS measurements of the ICG concentration in the cortex, and Ca(t) was measured by a dye densitogram unit attached to a forefoot of the piglet (model DDG-2001 A/K, Nihon Kohden, Tokyo, Japan). The tissue and blood ICG concentration curves were acquired over a 70-s sampling period. The impulse residue function, R(t), is the solution of Q(t) when Ca(t) is a delta function (37). The expression, CBF·R(t), was solved for by deconvolving Q(t) and Ca(t) (7). CBF and CBV were determined using the methodology of Zierler (37). Specifically, CBF is the initial height of the impulse residue function and CBV is determined from the area under the function.

CMRO2.   For a complete discussion on the use of NIRS to measure CMRO2, see Brown et al. (5). Briefly, measurements of CMRO2 were based on the Fick principle

Formula 4(4)
The arterial concentration of oxygen was determined by analyzing blood samples with a hemoximeter (Radiometer OSM3, Copenhagen, Denmark), and CBF was determined by the method explained above. The venous concentration of oxygen was determined indirectly from NIRS measurements of [Hb]. The total vascular concentration of Hb was determined by dividing [Hb] by the CBV. The venous Hb concentration was then calculated by assuming that CBV was composed of 20% arterial, 10% capillary, and 70% venous blood. The capillary concentration was assumed to be the average of the arterial and the venous concentrations (27).

The fraction of oxygen extracted from arterial blood into the brain, OEF, was calculated with the method described by Mintun et al. (20)

Formula 5(5)

Statistical analysis.   SPSS 10.1.0 was used for all statistical analyses. A repeated-measures two-way mixed ANOVA was used to compare measurements between the control and HI groups, with time as the within-subject variable and group as the between-subject variable. The analysis was conducted for all physiological parameters, CBF, CMRO2, and OEF. For parameters in which a two-way time-by-group effect was revealed by the omnibus test, group differences at individual time points were uncovered with a one-way ANOVA. Statistical significance was based on P < 0.05. All data are presented as means ± SE.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Table 1 summarizes the results of the measured physiological parameters for both control and HI piglets. Data are presented at baseline, for three time points during the insult, 5 min after the insult, and averaged over the last 3 h of the experiment. For the control group, no significant changes from baseline were observed in any physiological parameter throughout the study. During the insult, the HI group experienced an initial increase in mean arterial pressure, followed by a gradual decrease resulting in mean arterial pressure values below 70% baseline for at least the last 10 min of the insult. Arterial pH gradually decreased throughout the insult, reaching a low at 5 min after reoxygenation. Glucose and heart rate slowly increased during the insult, each peaking at 5 min after reoxygenation. All physiological parameters, excluding an elevated heart rate, returned to baseline in the last 3 h of the experiment.


View this table:
[in this window]
[in a new window]
  		Table 1. Physiological variables

 
Figure 1 presents the averaged CBF values for both the control group and the HI group during 30 min of baseline, for three time points during the insult, and for 6 h after the insult. There were no significant differences in baseline CBF between groups. The average baseline CBF was 70.9 ± 9.0 and 70.8 ± 5.1 ml·min–1·100 g–1 for control and HI groups, respectively. For piglets in the HI group, CBF dropped to 11 ± 5 ml·min–1·100 g–1 during the course of the insult and rebounded to 122 ± 15 ml·min–1·100 g–1 within 5 min of reoxygenation. The CBF of the control group did not change from baseline throughout the experiment. The pre- and postinsult data showed a weak time-by-group interaction [F(11,176) = 3.868, P < 0.05, power > 0.80]. The source of the interaction was the brief period of hyperemia experienced by the piglets in the HI group at 5 min after reoxygenation. There were no significant differences between groups at any other time points. Averaged over the last 3 h of the experiment, CBF was 66.9 ± 7.3 and 62.7 ± 10.3 ml·min–1·100 g–1 for control and HI groups, respectively.


Figure 1
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Cerebral blood flow (CBF) in control and hypoxia-ischemia (HI) piglets. Average CBF values throughout the study for control (solid line) and HI (dotted line) groups are shown. Gray area represents HI insult. Error bars are SE. *P < 0.05 between groups.

 
Figure 2 presents the averaged CBV values for both the control and HI groups during 30 min of baseline and for 6 h after the insult. The CBV values during the insult are not presented in Fig. 2 because they were underestimated in six of the nine piglets (see DISCUSSION). The average baseline value of CBV was 5.0 ± 0.3 and 5.1 ± 0.3 ml/100 g for control and HI groups, respectively. The data showed a weak time-by-group interaction [F(11,176) = 3.736, P < 0.05, power > 0.80]. The source of the interaction was the brief period of increased CBV experienced by the HI group at 5 min postinsult. At this time point, CBV was 6.9 ± 0.5 ml/100 g for HI piglets. There were no significant differences between groups at any other time points. The CBV of the control group did not change significantly from baseline throughout the experiment.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2. Cerebral blood volume (CBV) in control and HI piglets. Average CBV values throughout the study for control (solid line) and HI (dotted line) groups are shown. Gray area represents HI insult. Error bars are SE. *P < 0.05 between groups.

 
Figure 3 presents the averaged OEF values for both the control group and the HI group during 30 min of baseline and for 6 h after the insult. The OEF values during the insult are not presented in Fig. 3 because they were physiologically meaningless. The average OEF values at 5, 15, and 25 min during the insult were 1.29 ± 0.76, 3.71 ± 1.60, and 16.61 ± 10.31, respectively. The average baseline value of OEF was 0.30 ± 0.02 and 0.33 ± 0.03 for control and HI groups, respectively. The data showed a weak time-by-group interaction [F(11,176) = 3.399, P < 0.05, power > 0.80]. The source of the interaction was the brief period of depressed OEF experienced by the HI group at 5 min postinsult. At this time point, OEF was 0.15 ± 0.03 for HI piglets. There were no significant differences between groups at any other time points. The OEF of the control group did not change significantly from baseline throughout the experiment.


Figure 3
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 3. Oxygen extraction fracture (OEF) in control and HI piglets. Average OEF values throughout the study for control (solid line) and HI (dotted line) groups are shown. Gray area represents HI insult. Error bars are SE. *P < 0.05 between groups.

 
Figure 4 presents the averaged CMRO2 values for both the control group and the HI group during 30 min of baseline and for 6 h after the insult. Considering the error in the insult OEF values, CMRO2 measurements at the three time points during hypoxia ischemia are not shown. The data showed a strong time-by-group interaction [F(11,176) = 5.815, P < 0.001, power > 0.90]. The source of the interaction between the two groups was the decrease in CMRO2 experienced by the HI group after the insult that persisted for the remainder of the experiment. There were no statistical differences between groups at baseline. The average baseline value of CMRO2 was 2.18 ± 0.10 and 2.32 ± 0.11 ml O2·min–1·100 g–1 for control and HI groups, respectively. Immediately after the insult, the CMRO2 of the HI group dropped to 1.66 ± 0.12 ml O2·min–1·100 g–1 and remained depressed for the remainder of the study. The CMRO2 of the control group did not change significantly from baseline throughout the experiment.


Figure 4
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Cerebral metabolic rate of O2 (CMRO2) in control and HI piglets. Average CMRO2 values throughout the study for control (solid line) and HI (dotted line) groups are shown. Gray area represents HI insult. Error bars are SE. *P < 0.05 between groups.

 
Table 2 displays a comparison between CMRO2 and OEF measurements acquired with the NIRS technique and with the blood sample technique before and at 3 and 5 h after HI in six piglets. A repeated-measures ANOVA found no significant differences between the methods used to calculate CMRO2 at any of the three time points [F(2,20) = 0.321, P > 0.6, power < 0.10]. The NIRS technique was found to agree with the blood sample technique within a 13% error (P < 0.05, power > 0.8) at all time points. A time effect was found [F(2,20) = 21.891, P < 0.01, power > 0.99]. A paired t-test analysis was then applied to determine the source of the time effect. Measurements of CMRO2 acquired postinsult were reduced compared with preinsult measurements (P < 0.05), whereas measurements taken at the two postinsult time points were not significantly different from each other. This held true for both methods of calculating CMRO2. On average, CMRO2 dropped by 23.0 ± 4.9% after the insult. There was no difference between OEF values before and after the insult, and there was no difference between OEF values calculated from the two methods.


View this table:
[in this window]
[in a new window]
  		Table 2. Comparison of CMRO2 and OEF measurements acquired concurrently with NIRS and blood samples in 6 piglets at baseline (preinsult) and at 3 and 5 h after HI

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
A significant fraction of neuronal death associated with HI can be delayed by up to 6–24 h after resuscitation (10). Consequently, a key component to the effective treatment of brain damage resulting from a HI episode is the early detection of delayed brain injury. Reductions in cerebral oxidative metabolism after HI correlate with the severity of delayed brain injury (2). The objective of our group was to develop a clinically relevant NIRS technique capable of measuring these reductions at the bedside. By combining NIRS measurements of CBF with measurements of cerebral Hb concentration, we developed a NIRS technique for quantitatively measuring CMRO2 (4, 5, 34). A similar approach for measuring CMRO2 was recently published that used a rapid change in SaO2 as a blood-flow indicator (16). The primary advantage of using ICG as a tracer is the increased precision in the CBF measurements (23). As a prerequisite to clinical investigations, the purpose of the present study was to investigate whether the proposed technique is sensitive to reductions in oxidative metabolism after HI in piglets.

After a 30-min HI episode, the NIRS measurements of CMRO2 were consistently lower than baseline and lower than the corresponding CMRO2 values collected in sham-operated animals. The physiological parameters used to determined CMRO2 (see Eq. 3), namely CBF, CBV, [Hb], and SaO2, were not significantly different from controls after the first 30 min of reoxygenation. This indicates that the HI-related changes in CMRO2 do not alter CBF, CBV, or [Hb] preferentially. Rather, the change in CMRO2 was accommodated by a combination of smaller, and in these experiments statistically insignificant changes in CBF, CBV, and [Hb]. This implies that the parameters used to calculate CMRO2 are not sensitive enough to be used independently to detect HI-related reductions in oxidative metabolism, which highlights the need to calculate CMRO2 for diagnostic purposes.

The CMRO2 results of this study agree with at least two other studies that measured CMRO2 invasively before and after HI in newborn lambs. The study by Shadid et al. (31) reported a 35% reduction in CMRO2 at 60 min after HI. The study by Rosenberg (29) reported that both CMRO2 and CBF dropped by 28% after HI. The decrease in CMRO2 correlates well with our results; however, we did not measure a corresponding decrease in CBF. This difference may indicate that the HI insult was more severe than that used in our study. A reduction in CBF after a prolonged 75-min insult was also demonstrated by Okubo et al. (22). Impaired oxidative metabolism after HI has also been indirectly demonstrated using 31P-MRS to measure a decrease in the phosphocreatine-to-inorganic phosphate ratio. Using a piglet model with an insult of similar severity to that used in the present study, Kusaka et al. (17) measured a 30% decrease in this ratio at 3 and 6 h postinsult. In another 31P-MRS study, Peeters-Scholte et al. (25) reported a 13% drop in the ATP to exchangeable phosphate pool ratio at 1 and 2 h after HI.

Not all studies investigating oxidative metabolism after HI agreed with our results. Solås et al. (32) reported no significant changes in CMRO2 or CBF after HI in piglets within the first 2 h of reoxygenation. Their data appear to suggest a trend of reduced CMRO2 following HI, although this trend was not significant. Chang et al. (8) used NIRS to measure changes in cytochrome oxidase concentration after HI, which is an indirect measure of oxidative metabolism, and found no significant decrease compared with controls until ~30 h postinsult. The varying changes in oxidative metabolism measured by these studies highlight the potential for using CMRO2 measurements as a method of predicting the severity of delayed brain injury since the change in energy metabolism after the insult appears to be sensitive to the severity of the insult. We plan to investigate this dependency by measuring CMRO2 after varying levels of HI injury.

The NIRS measurements of CMRO2 are based on two assumptions. First, the concentration of water in the brain is 85%. Second, the ratio of arterial to venous blood relative to the total CBV remains constant at 25/75. We have recently validated the NIRS measurements of CMRO2 in the normal brain over a range from 1.5 to 4.0 ml O2·min–1·100 g–1 (34). However, it is possible that the assumed values for the water concentration and the arterial-to-venous ratio are altered by HI. The effect of cerebral edema after HI causes an increase in the brain water content of ~1% during the first 24 h postinsult (28), which indicates that this source of error is minimal. To verify that the proposed NIRS method can accurately measure CMRO2 following HI, we independently measured CMRO2 using blood samples to determine the arterial-venous O2 difference. The results presented in Table 2 demonstrate that both techniques measured a reduction in CMRO2 after HI; furthermore, there was no statistical difference in the postinsult CMRO2 values determined by the two techniques at 3 and 5 h postinsult. The agreement between the CMRO2 measurements from the two techniques demonstrates that the assumptions regarding water content and arterial-to-venous blood ratio are not significantly altered after HI.

Although the validation was conducted at only two time points following the HI insult (3 and 5 h), the constancy of the NIRS CMRO2 measurements throughout the 6-h postinsult period suggests that the validation holds for all time points. This is also reflected in the hemodynamic data. The CBF and CBV data did not differ statistically from baseline at any postinsult time point except for the 5-min mark. This indicates that changes in the arterial-to-venous ratio, such as caused by arteriole vasodilation, were unlikely because these changes would be reflected in CBF and CBV. At 5 min, there was a brief period of hyperemia (CBF reached 122 ± 15 ml·min–1·100 g–1), which could cause a change in the arterial-to-venous ratio. However, the CMRO2 value measured at this time was not statistically different from the other postinsult measurements. Furthermore, in our previous study, excellent agreement was found between CMRO2 measurements from the NIRS and blood sampling methods at similar hyperemic levels, indicating that there was no significant change in the vascular ratio (34).

A validation of the NIRS CMRO2 measurements during the HI insult was not attempted because of the difficulties associated with extracting a sufficient sample of blood from the sagittal sinus catheter under ischemic conditions. Even without this validation, it was evident that the NIRS measurements of CMRO2 during HI were incorrect based on the unrealistic values of OEF collected during this time. One cause of the spurious NIRS CMRO2 measurements was that the data acquisition period after the ICG injection was not long enough to properly characterize the impulse residue function (Eq. 3). The original acquisition time (70 s) was chosen from our previous studies (4, 5). During the course of these experiments, we discovered that this duration was not sufficient to account for the increased vascular transit time during ischemia. This caused the area under the impulse residue function to be truncated, which in turn resulted in the underestimation of CBV in six of the nine piglets. Because the vascular concentration of Hb was determined by dividing the NIRS measurement of total [Hb] by CBV, an underestimation in CBV causes an overestimation in both OEF and CMRO2. In the remaining three piglets, increasing the acquisition period to 120 s eliminated this problem.

Even if the acquisition time was increased to accurately determine CBV, our results indicate that OEF measured during HI was still not realistic. The average OEF in the three experiments in which CBV was measured properly was –0.95 ± 0.45, which is outside of the physiologically meaningful range. The inability to accurately define OEF, and subsequently CMRO2, is primarily caused by increased sensitivity to uncertainties in the CBV and [Hb] measurements. During HI, the difference between the arterial and venous oxygen concentrations is considerably smaller than under normal conditions. As a consequence, any measurement error in the venous concentration, which is proportional to [Hb]/CBV, is magnified in the calculations of OEF and CMRO2. This enhanced noise sensitivity is illustrated in Fig. 5, A and B, and discussed in more detail in the APPENDIX.


Figure 5
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 5. Sensitivity of OEF to changes in deoxyhemoglobin concentration ([Hb]; A) and CBV (B) for both normal (dashed line) and HI (solid line) conditions. [Hb] range was normalized to the mean [Hb] value under each condition ± 30% of the mean. For HI, a change of 30% is equivalent to 0.75 times the standard deviation of the measured data. Under normal conditions, a change of 30% is equivalent to 1.6 times the standard deviation.

 
A change in the arterial-to-venous ratio during the insult could also affect the OEF and CMRO2 measurements. For example, a change from 25/75 to 50/50, which has been observed under severe hypoxic conditions (15), would cause both OEF and CMRO2 to be underestimated by 30%. This effect, however, is too small to account for the magnitude of the observed errors. Furthermore, changing the arterial-to-venous ratio to 50/50 would only increase the error because the calculated OEF is already negative. An alternate and more reasonable estimation of CMRO2 can be obtained by assuming that the OEF during the insult is 1. In this case, CMRO2 is equal to the rate of oxygen delivery, which is defined as the product of CBF and SaO2. Both of these parameters can be measured accurately during the insult. It is important to remember that, although the NIRS measurements of CMRO2 were incorrect during the insult, the purpose of this study was to investigate the ability of NIRS to detect depressed oxidative metabolism within 6 h after resuscitation, since reduced CMRO2 during the recovery period could be a predicator of delayed brain injury (14).

In this study, we have demonstrated that NIRS can be used to measure changes in oxygen consumption after HI. This method has the potential to improve the diagnosis and treatment of HI injury for a number of reasons. First, considering the brevity of the therapeutic window, detecting early signs of brain injury could assist in the choice of therapeutic strategy for individual patients. Second, because neonates at risk of brain damage are often critically ill, the ability to measure CMRO2 at the bedside would allow frequent monitoring of treatment progress during the critical first few hours after HI. Future experiments comparing insult severity, in combination with tissue staining for apoptosis, with NIRS measurements of CMRO2 would be beneficial. This will assess the predictive power of the NIRS technique, which is a critical step toward implementing this technique in the neonatal intensive care unit.


   APPENDIX
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Sensitivity of OEF to [Hb] and CBV Under Normal and HI Conditions

The purpose of this appendix is to demonstrate the sensitivity of the calculated OEF to changes in the two NIRS parameters: [Hb] and CBV. The OEF was calculated under normal and HI conditions from Eq. 5 using the average parameter values measured in this study. To determine the arterial-venous difference in oxygen, the venous O2 concentration was calculated from the NIRS measurement of [Hb] using (5, 34)

Formula 6(6)
where [O2]v is the venous O2 concentration, [tHb] is the total hemoglobin concentration, [Hb]a is the arterial Hb concentration, {rho} is the density of brain tissue (1.05 g/ml), and 1.39 is the oxygen carrying capacity of hemoglobin measured in milliliters O2 per gram of Hb. The average [Hb] values were obtained from Table 1, namely, 60 ± 12 and 15 ± 1 µM under HI and normal conditions, respectively (mean ± SE). The average CBV values under baseline and HI conditions were 5.1 ± 0.2 and 6.5 ± 0.9 ml blood/100 g, respectively (mean ± SE). The latter was determined from the three experiments in which the acquisition time was sufficient to properly determine CBV.

Figure 5A displays the sensitivity of OEF to [Hb] under the two conditions. While all other parameters were kept constant, OEF was calculated over a range of [Hb], which was defined as the mean [Hb] for that condition ± 30%. For HI, the range was from 42 to 78 µM, which is 0.5 times the standard deviation of the measured data. Under normal conditions, the range was from 10 to 19 µM, which is 1.5 times the standard deviation. Figure 5B displays the sensitivity of OEF to CBV under the two conditions. In both cases, OEF was calculated over the CBV range from 2 to 8 ml blood/100 g.

The results presented in Fig. 5 show that the sensitivity of OEF to either [Hb] or CBV significantly increases during HI conditions. Small changes in either parameter can lead to large changes in the calculated OEF that can quickly exceed the physiological range. For comparison, the average experimental OEF values at 5, 15, and 25 min during the insult were 1.29 ± 0.76, 3.71 ± 1.60, and 16.61 ± 10.31, respectively. These simulations demonstrate the limitation of using the combined NIRS measurements of [Hb] and CBV to determine OEF under extreme conditions such as during HI. Because CMRO2 is given by CBF x OEF x arterial [O2], as derived from Eqs. 4 and 5, the calculated CMRO2 from NIRS would have equivalent errors. This analysis focused on OEF because it is easier to visualize how the enhanced sensitivity leads to nonphysiological values.


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
ACKNOWLEDGMENTS
 REFERENCES
 
This research was supported by the Canadian Institutes of Health Research.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
REFERENCES
 
We thank Y. Bureau for help with data analysis. We also thank D. Ouimet and V. Lun for assistance with the experiments.

Present address of D. W. Brown: University Hospital of Zurich, Zurich, Switzerland.


   FOOTNOTES
 

Address for reprint requests and other correspondence: K. St. Lawrence, Imaging Division, Lawson Health Research Institute, 268 Grosvenor St., London, Ontario, Canada N6A 4V2 (e-mail: kstlaw{at}lawsonimaging.ca)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Amato M and Donati F. Update on perinatal hypoxic insult: mechanism, diagnosis and interventions. Eur J Paediatr Neurol 4: 203–209, 2000.[CrossRef][Medline]
  2. Blumberg RM, Cady EB, Wigglesworth JS, McKenzie JE, and Edwards AD. Relation between delayed impairment of cerebral energy metabolism and infarction following transient focal hypoxia-ischemia in the developing brain. Exp Brain Res 113: 130–137, 1999.
  3. Boas DA, Gaudette T, Strangman G, Cheng X, Marota JJA, and Mandeville JB. The accuracy of near infrared spectroscopy and imaging during focal changes in cerebral hemodynamics. Neuroimage 13: 76–90, 2001.[ISI][Medline]
  4. Brown DW, Picot PA, Naeini JG, Springett R, Delpy DT, and Lee TY. Quantitative near infrared spectroscopy measurement of cerebral hemodynamics in newborn piglets. Pediatr Res 51: 564–570, 2002.[ISI][Medline]
  5. Brown DW, Hadway J, and Lee TY. Near infrared spectroscopy measurement of oxygen extraction fraction and cerebral metabolic rate of oxygen in newborn piglets. Pediatr Res 54: 861–867, 2003.[CrossRef][ISI][Medline]
  6. Cady EB, Costello AM, Dawson MJ, Delpy DT, Hope PL, Reynolds EO, Tofts PS, and Wilkie DR. Non-invasive investigation of cerebral metabolism in newborn infants by phosphorous nuclear magnetic resonance spectroscopy. Lancet 1: 1059–1062, 1983.
  7. Cenic A, Nabavi DG, Craen RA, Gelb AW, and Lee TY. Dynamic CT measurement of cerebral blood flow: a validation study. Am J Neuroradiol 20: 63–73, 1999.[Abstract/Free Full Text]
  8. Chang YS, Park WS, Lee M, Kim KS, Shin SM, and Choi JH. Near infrared spectroscopic monitoring of secondary cerebral energy failure after transient global hypoxia-ischemia in the newborn piglet. Neurol Res 21: 216–224, 1999.[ISI][Medline]
  9. Cooper CE, Elwell CE, Meek JH, Matcher SJ, Wyatt JS, Cope M, and Delpy DT. The noninvasive measurement of absolute cerebral deoxyhemoglobin concentration and mean optical path length in the neonatal brain by second derivative near infrared spectroscopy. Pediatr Res 39: 32–38, 1996.[ISI][Medline]
  10. Du Plessis AJ and Volpe JJ. Perinatal brain injury in the preterm and term newborn. Curr Opin Neurol 15: 151–157, 2002.[CrossRef][ISI][Medline]
  11. Frøen JF, Munkeby BH, Stray-Pedersen B, and Saugstad OD. Interleukin-10 reverses acute detrimental effects of endotoxin-induced inflammation on perinatal cerebral hypoxia-ischemia. Brain Res 942: 87–94, 2002.[CrossRef][Medline]
  12. Gillard E and Hagberg H. NMDA receptor-dependent increase of cerebral glucose utilization after hypoxia-ischemia in the immature rat. J Cereb Blood Flow Metab 16: 1005–1013, 1996.[CrossRef][Medline]
  13. Gillard E, Puka-Sundvall M, Hillered L, and Hagberg H. Mitochondrial function and energy metabolism after hypoxia-ischemia in the immature rat brain: involvement of NMDA-receptors. J Cereb Blood Flow Metab 18: 297–304, 1998.[CrossRef][ISI][Medline]
  14. Hagberg H. Mitochondrial impairment in the developing brain after hypoxia-ischemia. J Bioenerg Biomembr 36: 369–373, 2004.[CrossRef][ISI][Medline]
  15. Hueber DM, Franceschini MA, Ma HY, Zhang Q, Ballesteros JR, Fantini S, Wallace D, Ntziachristos, and Chance B. Non-invasive and quantitative near-infrared haemoglobin spectrometry in the piglet brain during hypoxic stress, using a frequency-domain multidistance instrument. Phys Med Biol 46: 41–62, 2001.[CrossRef][ISI][Medline]
  16. Kissack CM, Garr R, Wardle SP, and Weindling AM. Cerebral fractional oxygen extraction is inversely correlated with oxygen delivery in the sick, newborn, preterm infant. J Cereb Blood Flow Metab 25: 545–553, 2005.[Medline]
  17. Kusaka T, Matsuura S, Fujikawa Y, Okubo K, Kawad K, Namba M, Okada H, Imai T, Isobe K, and Itoh S. Relationship between cerebral interstitial levels of amino acids and phosphorylation potential during secondary energy failure in hypoxic-ischemic newborn piglets. Pediatr Res 55: 273–279, 2004.[Medline]
  18. Lorek A, Takei Y, Cady EB, Wyatt JS, Penrice J, Edwards AD, Peebles D, Wylezinska M, Owen-Reece H, Kirkbride V, Cooper CE, Aldridge RF, Roth SC, Brown G, Delpy DT, and Edwards EOR. Delayed ("secondary") cerebral energy failure after acute hypoxia-ischemia in the newborn piglet: continuous 48-hour studies by phosphorus magnetic resonance spectroscopy. Pediatr Res 36: 699–706, 1994.[ISI][Medline]
  19. Matcher SJ, Cope M, and Delpy DT. Use of the water absorption spectrum to quantify tissue chromophore concentration changes in near-infrared spectroscopy. Phys Med Biol 39: 177–196, 1994.[CrossRef][ISI][Medline]
  20. Mintun MA, Raichle ME, Martin WR, and Herscovitch P. Brain oxygen utilization measured with O-15 radiotracers and positron emission tomography. J Nucl Med 25: 177–187, 1984.[Abstract/Free Full Text]
  21. Mishra OP and Delivoria-Papadopoulos M. Cellular mechanisms of hypoxic injury in the developing brain. Brain Res 48: 233–238, 1998.
  22. Okubo K, Imai T, Namba M, Kusaka T, Yasuda S, Kawada K, Isobe K, and Itoh S. Estimation of cerebral blood flow in a newborn piglet model of neonatal asphyxia. Adv Exp Med Biol 540: 7–11, 2003.[Medline]
  23. Patel J, Marks K, Roberts I, Azzopardi D, and Edwards AD. Measurement of cerebral blood flow in newborn infants using near infrared spectroscopy with indocyanine green. Pediatr Res 43: 34–39, 1998.[ISI][Medline]
  24. Peeters C and van Bel F. Pharmacotherapeutical reduction of post-hypoxic-ischemic brain injury in the newborn. Biol Neonate 79: 274–280, 2001.[CrossRef][ISI][Medline]
  25. Peeters-Scholte C, Koster J, Veldhuis W, van den Tweel E, Zhu C, Kops N, Blomgren K, Bär D, van Buul-Offers S, Hagberg H, Nicolay K, van Bel F, and Groenendaal F. Neuroprotection by selective nitric oxide synthase inhibition at 24 h after perinatal hypoxia-ischemia. Stroke 33: 2304–2310, 2002.[Abstract/Free Full Text]
  26. Penrice J, Lorek A, Cady EB, Amess PN, Wylezinska M, Cooper CE, D'Souza P, Brown GC, Kirkbride V, Edwards AD, Wyatt JS, and Reynolds EO. Proton magnetic resonance spectroscopy of the brain during acute hypoxia-ischemia and delayed cerebral energy failure in the newborn piglet. Pediatr Res 41: 795–802, 1997.[ISI][Medline]
  27. Phelps ME, Huang SC, Hoffman EJ, and Kuhl DE. Validation of tomographic measurement of cerebral blood volume with C-11-labeled carboxyhemoglobin. J Nucl Med 20: 328–334, 1979.[Abstract/Free Full Text]
  28. Qiao M, Latta P, Meng S, Tomanek B, and Tuor UI. Development of acute edema following cerebral hypoxia-ischemia in neonatal compared with juvenile rats using magnetic resonance imaging. Pediatr Res 55: 101–106, 2004.[Medline]
  29. Rosenberg AA. Cerebral blood flow and O2 metabolism after asphyxia in neonatal lambs. Pediatr Res 20: 778–782, 1986.[Medline]
  30. Rosenberg AA, Parks JK, Murdaugh E, and Parker WD Jr. Mitochondrial function after asphyxia in newborn lambs. Stroke 20: 674–679, 1989.[Abstract]
  31. Shadid M, Moison R, Steendijk P, Hiltermann L, Berger HM, and van Bel F. The effect of antioxidative combination therapy on post hypoxic-ischemic perfusion, metabolism, and electrical activity of the newborn brain. Pediatr Res 44: 119–124, 1998.[ISI][Medline]
  32. Solås AB, Munkeby BH, and Saugstad OD. Comparison of short- and long-duration oxygen treatment after cerebral asphyxia in newborn piglets. Pediatr Res 56: 125–131, 2004.[CrossRef][ISI][Medline]
  33. Springett R, Sakata Y, and Delpy DT. Precise measurement of cerebral blood flow in newborn piglets from the bolus passage of indocyanine green. Phys Med Biol 46: 2209–2225, 2001.[CrossRef][ISI][Medline]
  34. Tichauer KM, Hadway J, Lee TY, and St. Lawrence K. Near-infrared spectroscopy measurement of cerebral oxidative metabolism: a validation study. J Cereb Blood Flow Metab First published September 28, 2005; doi:10.1038/sj.jcbfm.9600230.
  35. Van Bel F and Peeters C. Pharmacotherapeutical reduction of post-hypoxic-ischemic brain injury in the newborn. Biol Neonate 79: 274–280, 2001.
  36. Veenhoven RH, Vande Walle JG, Donckerwolcke RA, Wit JM, Griffiven AW, Derkx FH, and Schalekamp MA. Spectroscopic imaging in full term infants with perinatal asphyxia (Abstract 2009). Eleventh Annual Meeting: Society of Magnetic Resonance Medicine, 1992.
  37. Zierler KL. Equations for measuring blood flow by external monitoring of radioisotopes. Circ Res 16: 309–321, 1965.[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
100/3/850    most recent
00830.2005v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tichauer, K. M.
Right arrow Articles by Lawrence, K. St.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tichauer, K. M.
Right arrow Articles by Lawrence, K. St.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.