Effect of Acetazolamide on Regional Cerebral Oxygen Saturation and Regional Cerebral Blood Flow
Background and Purpose To verify whether the monitoring of regional cerebral oxygen saturation (rSO2) with transcranial near-infrared spectroscopy would successfully reflect changes in intracranial hemodynamics but not changes in extracranial compartment, we measured rSO2 and regional cerebral blood flow (rCBF) simultaneously in seven patients with cerebral ischemia and five normal volunteers before and after acetazolamide administration.
Summary of Report The baseline values of rSO2 and rCBF were 64.2±5.6% and 53.9±11.1 mL/100 g per minute, respectively. rCBF increased by 44.4±23.3% and rSO2 significantly increased to 69.6±5.6% after acetazolamide administration. Bilateral simultaneous measurement of rSO2 indicated a tendency that the larger the ΔrSO2, the greater the Δ%rCBF. The relationship between rSO2 level and rCBF value fit significantly on the theoretical curve calculated from Fick’s equation.
Conclusions It is suggested that monitoring of rSO2 with INVOS-3100 could be a useful indicator in the evaluation of intracranial hemodynamic changes.
As a simple but reliable method to evaluate in-tracerebral oxygenation, transcranial NIRS has become widely used to monitor cerebral hemodynamics and oxygenation.1 2 3 4 However, conflicting data regarding the accuracy of this method were recently presented by Harris and Bailey,5 indicating that hypercapnia induced during general anesthesia made a very small increase in rSO2 despite the more than doubling of the middle cerebral artery flow velocity. They speculated that external carotid flow might severely affect the results of NIRS.5 So far, to clarify this problem, few studies have directly correlated the change in rSO2 with the change in rCBF after hemodynamic alterations.
Since the intracranial microvasculature consists of approximately 75% venous, 20% arterial, and 5% capillary blood,6 7 the oximeter reading is weighted toward venous blood oxygen saturation, representing oxygen extraction by the cerebral tissue. When it is assumed that the cerebral vascular bed is 75% venous with negligible capillary volume, rSO2 could be represented as
AVDO2 was also expressed as
The nonlinear relationship between CBF and rSO2 illustrated in Fig 1⇓ was obtained from Equation 4.
In this study, we measured rSO2 and rCBF on the both sides of the patient’s forehead simultaneously before and after the administration of ACZ, which was the potential vasodilator10 verifying whether rSO2 measured with NIRS would successfully reflect changes in intracranial hemodynamics but not changes in the extracranial compartment.
Subjects and Methods
The measurement of rSO2 and rCBF was carried out in seven patients (mean age, 52±21 years; range, 14 to 70 years) with ischemic cerebrovascular disease and five normal volunteers (48±22 years; 23 to 73 years). CT scan demonstrated no ischemic changes in any patients in the frontal cortex where rSO2 monitoring was performed. There were two cases of unilateral ICA occlusion, two cases of unilateral ICA stenosis, one case of unilateral ICA occlusion with contralateral ICA stenosis, and one case of moyamoya disease. rSO2 was monitored simultaneously on both sides of the forehead with two cerebral oximeters (INVOS-3100; Somanetics Corp). rCBF was measured with a 133Xe intravenous injection technique (Valomet-1400). This method consisted of a bolus injection of 370 MBq 133Xe into a convenient arm vein followed by flushing with normal saline solution. The initial slope index11 was used for expressing rCBF. Fourteen NaI-collimated scintillation probes were applied over each cerebral hemisphere. Of the 14 probes, two probes were applied over each side of the forehead where the symmetrical rSO2 monitoring was carried out. The mean rCBF values of these two probes were used in this study.
After the baseline measurement of rCBF, rSO2 sensors (9×4 cm) were symmetrically placed on both sides of the forehead. After rSO2 levels were confirmed to be stable, 1 g ACZ was injected intravenously. As rSO2 reached plateau (usually 10 to 15 minutes later), the rCBF measurement was repeated.
The side-to-side asymmetry in increases of rSO2 and in the percentage increase of rCBF after ACZ challenges were expressed as AIs of rSO2 and rCBF, respectively. The following formulas were applied in this study:
where AS indicates affected side and US, unaffected side.
In normal volunteers, the rSO2 value and rCBF value of the left side were grouped with the affected side and those of the right side with the unaffected side, respectively.
The statistical values in this article are expressed as mean±SD. The rSO2 levels and rCBF values before and after ACZ administration were compared with two-tailed Wilcoxon signed-rank test. Correlations between ΔrSO2 and Δ%rCBF, between AI(rSO2) and AI(rCBF), and between rSO2 and 1/rCBF were evaluated by a simple regression analysis. A significant difference in the statistical results was defined as P<.05.
The baseline values of rSO2 and rCBF were 64.2±5.6% (range, 56.0% to 75.0%) and 53.9±11.1 mL/100 g per minute (range, 38.8 to 75.5 mL/100 g per minute), respectively. After administration of ACZ, rSO2 rose significantly to 69.6±5.6% within 15 minutes (P<.001). The increase in rSO2 was 5.4±3.2% (range, 0% to 14%). The percentage increase in rCBF was 44.4±23.3% (range, −2.0% to 109.4%) (Fig 2⇓). A significant linear regression was obtained between ΔrSO2 and Δ%rCBF (ΔrSO2=2.15+0.07 · Δ%rCBF; r=.521, P<.01). The AI(rSO2) was −37.1±61.8% (range, −200.0% to 18.2%), which significantly correlated with the AI(rCBF) of −29.5±41.8% (range, −127.3% to 47.0%) (Fig 3⇓, P<.02).
The relationship between rSO2 and rCBF before and after ACZ is illustrated in Fig 4⇓. Equation 4 was rewritten as
In this study, a significant rise of rSO2 was observed after ACZ administration, and the results (Figs 2⇑ and 3⇑) indicated a relationship whereby the greater the Δ%rCBF, the larger the ΔrSO2. ACZ, a selective inhibitor of carbonic anhydrase, has been shown to increase CBF markedly without any change of CMRO2.10 We did not analyze blood gases during the ACZ challenge; however, it has also been reported that ACZ administration significantly increases jugular venous oxygen saturation but does not alter arterial oxygen saturation.10 Furthermore, with an advanced MRI technique, it was recently demonstrated that ACZ administration induces cerebral venous hyperoxygenation in the cortical and subcortical gray matter.12 These findings suggest that a rise in CBF induced by ACZ directly elevates tissue oxygenation and venous oxygen saturation, which results in a rise of rSO2 level as monitored with an oximeter. From the theoretical relationship between rSO2 and rCBF depicted in Fig 1⇑, ΔrSO2/Δ%rCBF was considered to depend on two factors: rCBF level before ACZ administration and the Δ%rCBF value itself. Therefore, a simple analysis could not be applied to the relationship between ΔrSO2 and Δ%rCBF. However, from Fig 2⇑ at least, it was indicated that elevation of rSO2 was always induced where rCBF was increased after ACZ administration and that there was a relationship between them whereby the greater the Δ%rCBF, the larger the ΔrSO2.
Harris and Bailey,5 using the INVOS system, demonstrated that hypercapnia induced during general anesthesia made no significant increase in rSO2 despite the more than doubling of middle cerebral artery flow velocity. They speculated that the INVOS system reflected external carotid flow with minimal contribution from the internal carotid circulation. However, in this study, Fig 3⇑ demonstrated that AI(rSO2) significantly correlated with AI(rCBF) in wide ranges of rCBF asymmetry, which also implied that rSO2 measurement with INVOS-3100 would indicate the change of intracranial hemodynamics.
By use of Fick’s equation, the theoretical relationship between rSO2 and rCBF was obtained as Equation 4. Fig 5⇑, illustrating the relationship between rSO2 and 1/rCBF, demonstrated that rSO2 significantly reflected the hemodynamic changes that accompanied the ACZ challenges. No changes in ischemia were detected with CT scan where the measurement of rSO2 was carried out. However, CMRO2 was not calculated in this study. The individual differences in CMRO2 among subjects might be one of the main reasons that the statistical significance was not so strong and the intercept of the y axis in Fig 5⇑ was smaller than SAO2.
It is conclusively demonstrated in this study that rSO2 measurement with INVOS-3100 is a simple but useful method to evaluate cerebral hemodynamic changes.
Selected Abbreviations and Acronyms
|AVDO2||=||arteriovenous oxygen difference|
|CMRO2||=||cerebral metabolic rate of oxygen|
|ICA||=||internal carotid artery|
|rCBF||=||regional cerebral blood flow|
|rSO2||=||regional cerebral oxygen saturation|
- Received May 17, 1995.
- Revision received August 28, 1995.
- Accepted August 28, 1995.
- Copyright © 1995 by American Heart Association
Dujovny M, Slavin KV, Hernandez G, Geremia GK, Ausman JI. Use of cerebral oximetry to monitor brain oxygenation reserves for skull base surgery. Skull Base Surgery. 1994;4:117-121.
Mchedlishvili GI. Arterial behavior and blood circulation in the brain. New York, NY: Plenum Publishing Corp; 1986:56-57.
Vorstrup S, Henriksen L, Paulson OB. Effect of acetazolamide on cerebral blood flow and cerebral metabolic rate for oxygen. J Clin Invest. 1984;74:1634-1639.
Risberg J, Ali Z, Wilson EM, Halsey JH. Regional cerebral blood flow by 133 Xenon inhalation: preliminary evaluation of an initial slope index in patients with unstable flow compartments. Stroke. 1975;6:142-148.