Intensive care for patients with increased intracranial pressure still involves the use of moderate hyperventilation (1) to decrease cerebral blood volume (CBV), one determinant of intracranial pressure. However, the vasoconstrictive effect of hyperventilation is not maintained. Hyperventilation for 4–6 h results in a gradual increase in pial arteriolar diameter (2), global CBV (3,4) and cerebral blood flow (CBF) (5,6). Moreover, on termination of hyperventilation, CBF increases above control values (5,6). Most of these studies were performed with global measurements or in vitro techniques and gave no information on the regional CBV or CBF changes to sustained hypocapnia. The availability of new techniques make it possible to look for in vivo regional variations.
The aim of this study was to investigate the regional CBV response during 3 h of hyperventilation in thiopental-anesthetized, mechanically-ventilated rats to determine 1) the range and the evolution of CBV changes during hypocapnia and on termination of hyperventilation and 2) possible differences between brain areas in responsiveness to sustained hyperventilation. CBV was measured by using steady-state susceptibility contrast magnetic resonance imaging (MRI) (7–9) in dorsoparietal neocortex, corpus striatum, and cerebellum.
Susceptibility contrast MRI exploits the increase in the magnetic susceptibility difference (Δχ) between the vascular and extravascular compartments induced by the presence of a long-lived contrast agent confined in the vascular bed. This increase in Δχ results in an increase (ΔR2∗) of the decay rate (R2∗ = 1/T2∗) of the nuclear magnetic resonance signal from water protons, which is proportional to regional CBV, as previously shown (7,10): where γ is the gyromagnetic ratio, rCBV is the regional CBV, and B0 is the magnetic field in the absence of sample.
Nine fed Sprague-Dawley female rats (220–240 g) were studied. Preparation of animals conformed to the guidelines of the French Government (decree 87-848, October 19, 1987; license 006683). Anesthesia was induced with 4% halothane and maintained with 1% halothane during surgical manipulation. One percent lidocaine was injected subcutaneously for local anesthesia at all surgical sites. After tracheotomy, the rats were mechanically ventilated with 65% nitrous oxide/35% oxygen by using a rodent ventilator (Model 683; Harvard Apparatus Inc., South Natick, MA). Ventilation was adjusted to maintain Paco2 at ≈35 mm Hg. The fractional inspired oxygen was continuously monitored (MiniOX I analyzer; Catalyst Research Corp., Owings Mills, MD). A 0.7-mm indwelling catheter was inserted into the femoral artery to monitor mean arterial blood pressure (MAP). Blood gases (Pao2 and Paco2), arterial pH, and arterial hemoglobin (Hb) concentration were analyzed in arterial blood samples of <0.1 mL (ABL 510; Radiometer, Copenhagen, Denmark). Another 0.7-mm indwelling catheter was inserted into the left femoral vein for injection of the contrast agent. Rectal temperature was maintained at 37.0°C ± 0.5°C by using a heating pad placed under the abdomen. Blood gases and arterial pH were corrected for rectal temperature. After the end of surgical manipulation, inhaled anesthesia was changed to a continuous intraperitoneal administration of thiopental (40 mg · kg−1 · h−1) and pancuronium bromide (0.2 mg · kg−1 · h−1).
Animals were subjected to three successive episodes: normocapnia, 3-h hypocapnia, and recovery. The cycle was started after more than 30 min of equilibration from surgery. The initial exclusion criteria from the study were MAP <90 mm Hg, arterial pH <7.30, Pao2 <100 mm Hg, and arterial Hb <10 g/dL. The control period was normocapnia. Hyperventilation (Paco2 ≈ 25 mm Hg) was then induced by increasing the rate of minute ventilation, and it remained unchanged throughout the period. Finally, normocapnia was restored (recovery). When the cycle of measurements was ended, the animals were killed by the administration of 5% halothane in nitrous oxide.
A pilot study was conducted with a group of five rats to ascertain the stability of the Paco2 during 3 h of hyperventilation. As shown in Table 1, there was a good stability in blood gases and arterial pH during this procedure. These results allowed us to withdraw only three blood samples in the studied group (normocapnia, after 15 min of hypocapnia and recovery), to prevent anemia, hemodynamic instability, and a loss of the contrast agent because of repeated blood samples.
MRI was performed with a SMIS console (SMIS Ltd., Guildford, UK) equipped with a 2.35-T, 40-cm-diameter horizontal bore magnet (Bruker Spectrospin, Wissembourg, France) and a 20-cm-diameter actively shielded gradient (Magnex Scientific Ltd., Abdington, UK). The rat was prone, with its head secured via ear bars, and a 30-mm-diameter surface coil was located directly above the brain. After radiofrequency coil matching and tuning, the magnetic field homogeneity was adjusted to obtain a line width for water smaller than 0.2 parts per million (ppm) in the brain. Six adjacent horizontal slices (from 2 mm below the bregma) were chosen from T1 coronal, transverse, and sagittal scout images. A series of images for each slice at different echo times was acquired by using a multiple gradient echo sequence with an interecho interval of 8.28 ms (repetition time, 6000 ms; first echo time, 9.04 ms; field of view, 35 mm; slice thickness, 1 mm; 128 × 64 image matrix; number of echoes, 5; number of averages, 2). Acquisition of all images of the six slices lasted approximately 13 min. In the control period, images were acquired before and 3 min after IV injection of a superparamagnetic iron oxide agent (200 μmol of iron per kilogram of AMI-227, Sinerem®; Guerbet, Aulnay-sous-Bois, France). Then, acquisition of images at various echo times in the 6 slices was repeated 7 times during the hypocapnic episode. The repetition interval was 30 min. One series of images at various echo times in the six slices was acquired during the recovery episode.
Image postprocessing and determination of regional CBV were performed with an Ultrasparc workstation (Sun Microsystems, Pasadena, CA), as previously described (9). T2∗ images were calculated by a least-squares monoexponential fit of the intensity versus the echo time on a pixel-by-pixel basis. Differences in relaxation rates in each pixel were then calculated according to the formula MATH with T2∗pre and T2∗post being the decay time constants before and after administration of the contrast agent, respectively. The ΔR2∗ values were obtained from the T2∗ postcontrast values during normocapnia, hypocapnia, and recovery. Regions of interest were defined in the two hemispheres of the dorsoparietal neocortex, corpus striatum, and cerebellum. Selection of cerebral regions was made from the 3-mm-below-bregma slice (for neocortex) and from the 6-mm-below-bregma slice (for corpus striatum and cerebellum) by comparing the data with an anatomical atlas (11). Large ΔR2∗ values (>250/s) corresponding to a large blood volume fraction were discarded.
A correction for clearance of the contrast agent from the plasma was applied because the postcontrast experiments lasted approximately 210 min. The correction was based on the individual determination of the elimination half-life of the contrast agent. It was assumed that hyperventilation did not result in blood volume change of muscle tissue (12,13) and that the concentration of contrast agent versus time was monoexponential. A region of interest was chosen in extracerebral tissue (jaw muscle), and ln (ΔR2∗) of this region was plotted versus time (n = 9 repeated measurements) to calculate the elimination slope of the contrast agent by using a linear square regression. The calculated elimination half-life was then used for correcting measured intracerebral ΔR2∗ for each animal.
The regional CBV, expressed as a percentage of blood volume in each voxel, or milliliters per 100 g of tissue, was then determined according to Equation 1. For an injection of AMI-227 of 200 μmol of iron per kg of body mass, Δχ = 0.571 ppm at 2.35 T in large vessels (14). We assumed that the average hematocrit (Hct) in the brain microcirculation was 0.83 of that in large vessels (15,16), resulting in a Δχ value of 0.688 ppm. We assumed that the brain Hct remained constant during hypocapnia, as previously shown (16).
Data were expressed as mean ± sd. The differences between the two hemispheres were tested by using a paired Student’s t-test. Then pooled data were subjected to analysis. Analysis for statistical significance of changes during the successive episodes was performed with two-way (brain region × repeated episodes) or one-way analysis of variance for repeated measurements (StatView SE program; SAS Institute Inc., Cary, NC). Each value was compared with that obtained at the control period (normocapnia) by using the Scheffé F test as a post hoc test. The relation between mean CBV in each region and duration of hyperventilation was determined with linear square regression. Comparisons among the three brain regions were subjected to a factorial analysis of variance by using the Scheffé F test. Statistical significance was declared when P < 0.05.
Physiological data of the studied group (n = 9 rats) during the successive episodes are shown in Table 2. On termination of hyperventilation, there was a slight but significant decrease in MAP, arterial pH, and Hb content. However, no consistent change in repeated MAP measurements was found during the hyperventilation period.
The data used for the determination of elimination half-life of the contrast agent are shown in Figure 1. All experimental data were fitted according to a simple linear regression with coefficient correlation r2 ≥ 0.80. The elimination half-life was 5.1 ± 1.7 h.
No significant difference in CBV values was found between the two hemispheres. Baseline CBV was smaller in the corpus striatum (2.9 ± 0.6 mL/100 g) than in the neocortex (3.7 ± 0.8 mL/100 g) (P < 0.05). Cerebellum had an intermediate CBV value (3.3 ± 0.5 mL/100 g). At 15 min after the start of hypocapnia, regional CBV was significantly decreased: −16% ± 10% of control values in corpus striatum, −14% ± 9% in cerebellum, and −13% ± 8% in neocortex (all P < 0.05). However, sustained hypocapnia resulted in different responses in the three brain regions, as seen in Figure 2. In the corpus striatum, CBV remained stable during sustained hypocapnia, whereas a progressive increase in CBV was seen in neocortex and, to a lesser extent, in cerebellum. The relation between CBV changes in hypocapnia and the duration of hyperventilation resulted in different correlation coefficients between regions:r2 = 0.13 (P = 0.44) in corpus striatum, r2 = 0.83 (P = 0.02) in cerebellum, and r2 = 0.96 (P = 0.01) in neocortex. This trend was reflected by a significant interaction between brain regions (corpus striatum versus neocortex) and duration of hyperventilation (control, 15, 165, and 195 min) (F = 2.70;P = 0.05). Also, significant differences were found between corpus striatum and the other regions in CBV changes at 165 and 195 min after the start of hypocapnia (Table 3). All CBV changes returned to normal values in the recovery period, with no value significantly above control values; i.e., no evidence of rebound was found.
This is the first in vivo study that measured regional CBV during three hours of hyperventilation. Two new findings are reported here that may go beyond previous data from global CBV measurements. First, sustained moderate hyperventilation produced different CBV responses between the brain regions investigated, with the transient effect of sustained hypocapnia being more visible in neocortex and cerebellum than in corpus striatum. Second, no evidence of a rebound in CBV was found on recovery in the three brain regions, probably because of the moderate degree of hyperventilation.
A prerequisite for interpreting our data is the stability of the model. Changes in neither blood gases nor arterial pH were found throughout three hours of hypocapnia in a pilot study (see Table 1). From these results, we assumed that no further changes in Paco2 occurred in the studied group after 15 minutes of hypocapnia, because the minute ventilation was the same for three hours. The number of blood samples in the studied group was thus limited to prevent cerebrovascular changes and a loss of contrast agent because of repeated blood withdrawals. In addition, no consistent change in MAP was found during the hyperventilation period, in agreement with previous studies (6). Finally, a slight but significant arterial acidosis was observed during the recovery period, as previously reported by others (5,6). This acidosis is due to an incomplete restoration of the decreased blood concentration of bicarbonates induced by hyperventilation (6). It should be kept in mind that metabolic acidosis has no major effect on CBV (17). All these different points indicate that there was no major change in physiological data during three hours of moderate hyperventilation.
Steady-state susceptibility contrast imaging was successfully used to investigate cerebrovascular changes in brain during challenges such as hypercapnia or ischemia (8,9). Equation 1, used to determine absolute regional CBV, is based on a highly simplified model of the brain vessel architecture (10). Despite this approximation, the control values of regional CBV (2.9–3.7 mL/100 g) that we obtained are in reasonable agreement with other studies on rats that used different techniques (16,18).
The contrast agent clearance during the postcontrast experiment is an important issue. An individual correction for contrast agent clearance was applied by using ΔR2∗ versus time data in extracerebral tissue. The large molecular size of the contrast agent prevents it from leaving the vascular bed. In addition, hypocapnia does not significantly affect blood volume in extracerebral or muscle tissue (12,13). Using this approach, we determined elimination half-life values in reasonable agreement with the manufacturer’s data (4.5 hours; Guerbet laboratory, unpublished data). If we had used this value to uniformly correct the elimination of the contrast agent, the conclusions on regional CBV response to sustained hyperventilation would have been the same. However, the large individual variability (30%) in half-life values is in favor of an individual correction.
The response of CBV to sustained hyperventilation was clearly different among the three brain regions. The CBV of neocortex returned toward normal, and this gradual increase is in accordance with previous data. For example, the initial decrease in CBF and in CBV was followed by a return to the normal values within four to six hours of hyperventilation in goats (6), dogs (3,4), and humans (5). The reexpansion of cerebral vessels is believed to result from an adaptation of brain extracellular pH (2,19). At a constant level of hypocapnia, bicarbonate content in cerebrospinal fluid (CSF) decreases with time, probably through reduced secretion by the choroid plexus, and results in a return of CSF (and extracellular) pH to normal (6,20). These changes in brain CSF pH parallel those in CBF returning to normal prehypocapnic values (6,21). Because topical application of solution with low or high pH produces arteriolar dilation or constriction, respectively (22), a progressive change in extracellular pH could account for the cerebrovascular adaptation to sustained hypocapnia.
However, this overall mechanism is not a satisfactory explanation concerning this study. Such a regional difference was previously observed because the thalamus (which is included with the corpus striatum in the subcortex) was found to be the only brain region that did not recover its initial CBF in goats submitted to six hours of hyperventilation (6). Because the progressive change in extracellular pH is believed to have been similar in our three studied regions, one can wonder about underlying mechanisms that could explain a regional difference in brain vessel sensitivity to extracellular pH. This could be due to differences in blood volume; the smaller the baseline CBV, the slower the rate of change during sustained hyperventilation. Our baseline differences in CBV between corpus striatum and neocortex are in agreement with other data (15,23). In addition, the cerebrovascular response to acetazolamide was smaller in regions with small baseline CBV and small arterial-to-capillary blood volume (24). Similar results were found with regional CBF measurements during hypercapnia and hypocapnia (25). This indicates that the cerebrovascular response to CO2 alteration, and hence to change in extracellular pH, may depend on the baseline blood volume fraction reflecting the different numbers of vessels and collaterals. Another mechanism for this difference in regional CBV response could result from regional differences in brain vessel tone. There are several mediators that link extracellular pH and brain vascular tone (e.g., nitric oxide, cyclic nucleotides, potassium channels, and prostanoids), and all of these have vasodilating properties (26). Regional differences in neuronal nitric oxide synthase and in potassium channels have been recently reported in rats (27,28). However, whether these regional differences can be related to our results is unknown.
Our study of CBV was conducted in the presence of thiopental anesthesia. We might expect some effect of thiopental on the course of CBV during prolonged hyperventilation because thiopental decreases the response of cerebral arterial rings to hypocapnia (29). However, CBF changes induced by thiopental administration were found to be similar in corpus callosum, cerebellum, cerebral hemispheres, and brainstem during normocapnia versus hypocapnia (30). At 15 minutes of hypocapnia, the CBV reduction found in the three brain regions was close to that found in halothane-anesthetized rats during the same procedure (12). In dogs with an intracranial mass lesion, there were no differences in the reexpansion of global CBV among thiopental, fentanyl, and isoflurane during four hours of hyperventilation (4). Although any effect of anesthesia cannot be eliminated from our results, it is unlikely that continuous infusion of thiopental might have interfered with the progressive regional difference in response to hypocapnia. In addition, the reexpansion of neocortical CBV during sustained hyperventilation was not prevented by thiopental, confirming a previous study (4).
In conclusion, these findings indicate that sustained hyperventilation during thiopental anesthesia results in different regional CBV responses. The mechanism probably depends on the relationship between extracellular pH and regional CBV. If these results can be applied to clinical practice, they should indicate regional heterogeneity in the CBV response to three hours of hyperventilation, especially relevant during the management of patients with brain injury or ischemic stroke.
The authors thank Jonathan Coles and Richard Traystman for helpful comments on the manuscript.
1. Allen CH, Ward JD. An evidence-based approach to management of increased intracranial pressure. Crit Care Clin 1998; 14: 485–95.
2. Muizelaar JP, van der Poel HG, Li ZC, et al. Pial arteriolar vessel diameter and CO2
reactivity during prolonged hyperventilation in the rabbit. J Neurosurg 1988; 69: 923–7.
3. Artru AA. Reduction of cerebrospinal fluid pressure by hypocapnia: changes in cerebral blood volume, cerebrospinal fluid volume, and brain tissue water and electrolytes. J Cereb Blood Flow Metab 1987; 7: 471–9.
4. Artru AA. Reduction of cerebrospinal fluid pressure by hypocapnia: changes in cerebral blood volume, cerebrospinal fluid volume and brain tissue water and electrolytes. II. Effects of anesthetics. J Cereb Blood Flow Metab 1988; 8: 750–6.
5. Raichle ME, Posner JB, Plum F. Cerebral blood flow during and after hyperventilation. Arch Neurol 1970; 23: 394–403.
6. Albrecht RF, Miletich DJ, Ruttle M. Cerebral effects of extended hyperventilation in unanesthetized goats. Stroke 1987; 18: 649–55.
7. Boxerman JL, Hamberg LM, Rosen BR, Weisskoff RM. MR contrast due to intravascular magnetic susceptibility perturbations. Magn Reson Med 1995; 34: 555–66.
8. Hamberg LM, Boccalini P, Stranjalis G, et al. Continuous assessment of relative cerebral blood volume in transient ischemia using steady state susceptibility-contrast MRI. Magn Reson Med 1996; 35: 168–73.
9. Payen JF, Vath A, Koenigsberg B, et al. Regional cerebral plasma volume response to carbon dioxide using magnetic resonance imaging. Anesthesiology 1998; 88: 984–92.
10. Yablonskiy DA, Haacke EM. Theory of NMR signal behavior in magnetically inhomogeneous tissues: the static dephasing regime. Magn Reson Med 1994; 32: 749–63.
11. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. Sydney: Academic Press, 1982.
12. Payen JF, Briot E, Tropres I, et al. Regional cerebral blood volume response to hypocapnia using susceptibility contrast MRI. NMR Biomed 2000; 13: 384–91.
13. Hampson NB, Piantadosi CA. Near-infrared optical responses in feline brain and skeletal muscle tissues during respiratory acid-base imbalance. Brain Res 1990; 519: 249–54.
14. Tropres I, Grimault S, Vaeth A, et al. Vessel size imaging. Magn Reson Med 2001; 45: 397–408.
15. Cremer JE, Seville MP. Regional brain blood flow, blood volume, and haematocrit values in the adult rat. J Cereb Blood Flow Metab 1983; 3: 254–6.
16. Todd MM, Weeks JB, Warner DS. Microwave fixation for the determination of cerebral blood volume in rats. J Cereb Blood Flow Metab 1993; 13: 328–36.
17. Van de Ven MJ, Colier WN, van der Sluijs MC, et al. Ventilatory response in metabolic acidosis and cerebral blood volume in humans. Respir Physiol 2001; 124: 105–15.
18. Shockley RP, LaManna JC. Determination of rat cerebral cortical blood volume changes by capillary mean transit time analysis during hypoxia, hypercapnia and hyperventilation. Brain Res 1988; 454: 170–8.
19. Raichle ME, Plum F. Hyperventilation and cerebral blood flow. Stroke 1972; 3: 566–75.
20. Hornbein TF, Pavlin EG. Distribution of H+
minus between CSF and blood during respiratory alkalosis in dogs. Am J Physiol 1975; 228: 1149–54.
21. Pannier JL, Leusen I. Circulation to the brain of the rat during acute and prolonged respiratory changes in the acid-base balance. Pflugers Arch 1973; 338: 347–59.
22. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH, PCO2
and bicarbonate on pial vessels. Stroke 1977; 8: 358–60.
23. Bereczki D, Wei L, Otsuka T, et al. Hypercapnia slightly raises blood volume and sizably elevates flow velocity in brain microvessels. Am J Physiol 1993; 264: H1360–9.
24. Okazawa H, Yamauchi H, Sugimoto K, et al. Effects of acetazolamide on cerebral blood flow, blood volume, and oxygen metabolism: a positron emission tomography study with healthy volunteers. J Cereb Blood Flow Metab 2001; 21: 1472–9.
25. Ito H, Yokoyama I, Iida H, et al. Regional differences in cerebral vascular response to PaCO2
changes in humans measured by positron emission tomography. J Cereb Blood Flow Metab 2000; 20: 1264–70.
26. Brian JE Jr. Carbon dioxide and the cerebral circulation. Anesthesiology 1998; 88: 1365–86.
27. Stanarius A, Seidel B, Wolf G. Neuronal nitric oxide synthase in the vasculature of the rat brain: an immunocytochemical study using the tyramide signal amplification technique. J Neurocytol 1998; 27: 731–6.
28. Horiuchi T, Dietrich HH, Tsugane S, Dacey RG Jr. Role of potassium channels in regulation of brain arteriolar tone: comparison of cerebrum versus brain stem. Stroke 2001; 32: 218–24.
29. Ogawa K, Yamada S, Mizumoto K, et al. Inhibitory effects of halothane, isoflurane, sevoflurane, and pentobarbital on the constriction induced by hypocapnia and bicarbonate in isolated canine cerebral arteries. J Neurosurg Anesthesiol 2000; 12: 99–106.
© 2002 International Anesthesia Research Society
30. Kassell NF, Hitchon PW, Gerk MK, et al. Influence of changes in arterial PCO2
on cerebral blood flow and metabolism during high-dose barbiturate therapy in dogs. J Neurosurg 1981; 54: 615–9.