The number of patients with cerebral ischemia has been increasing recently, and this condition has become an important public health issue. Therefore, diagnosis and medical treatment for it are becoming increasingly important. Evaluation of cerebral perfusion is useful in patients with cerebral ischemia. Conventional methods for evaluation of cerebral hemodynamics currently include angiography, perfusion computed tomography (CT), perfusion magnetic resonance imaging (MRI), transcranial Doppler (TCD), single-photon-emission computed tomography (SPECT), and positron emission tomography (PET). These techniques allow the measurement of cerebral perfusion deficits, but they have several drawbacks. Angiography is invasive, perfusion CT and MRI require injection of a contrast agent, and SPECT and PET require administration of a radioisotope. TCD and arterial spin labeling MRI have been developed as noninvasive, in vivo imaging methods. However, with TCD, the measurement area is limited and the accuracy of the examination depends on the skill of the operator. Arterial spin labeling MRI is still insufficient with regard to diagnostic accuracy. All the aforementioned methods require that the patient be restrained, which makes repeated measurements difficult.12.3.–4 Furthermore, measurements at the bedside cannot be acquired because the equipment required for these examinations (other than TCD) are not easily transportable.
On the other hand, near-infrared spectroscopy (NIRS) can continuously measure real-time changes in the concentration of oxygenated hemoglobin (HbOxy), deoxygenated hemoglobin (HbR), and total hemoglobin (HbTotal) in the cerebral tissues.56.–7 HbTotal is defined as the sum of HbOxy and HbR. The basis of this approach is the modified Beer-Lambert law. This method is suitable for repeated measurement of changes in cerebral oxygenation at the patient’s bedside. Compared to other modalities, it is simple, noninvasive, requires minimal patient restraint, has high temporal resolution, and is transportable.8,9 Optical topography (OT) is a NIRS technique that provides spatial maps of hemodynamic and oxygenation changes.10 OT has been increasingly used in recent years in various clinical fields, including neurosurgery, neurology, and psychiatry.1112.–13
This report proposes a method for measuring cerebral oxygenation using OT with oxygen inhalation. This technique can be used to evaluate cerebral ischemia easily and repeatedly in real time and can be used to complement conventional methods.1415.–16
A total of 30 normal volunteers without a history of cerebral disease or risk factors of stroke (25 men, 5 women; age range, 22 to 56 years) and 33 patients with cerebral ischemia in whom the decrease in cerebral blood flow (CBF) in the area of the middle cerebral artery (MCA) was studied with N-isopropyl-p- Iodoamphetamine () SPECT (29 men, 4 women; age range, 58 to 78 years). Cases of stenosis or occlusion of the internal carotid artery (ICA) or MCA were visualized with MR angiography, and cerebral infarctions were not observed in any cases in the cerebral cortex in the area of the MCA. In patients with cerebral ischemia, the average duration from the time the vascular lesions were found by MR angiography to the OT measurement was 5.6 months. The lesions included 24 ICA stenoses, 4 ICA occlusions, 3 MCA stenoses, and 2 MCA occlusions.
This study was approved by the Ethics Committee of Jichi Medical University, and informed consent was obtained from all subjects.
Materials and Methods
Methods of Measurement for OT
Measurements were performed on a continuous-wave OT system (ETG-4000, Hitachi Medical Corporation, Tokyo, Japan) with 48 channels using two faces of probe arrays. In this system, near-infrared light rays with wavelengths of 695 and 830 nm were guided by optical fiber bundles and transmitted into the cranium. Reflections of the near-infrared rays were detected in receiving probes set on the scalp 30 mm away from the transmitting probes. Near-infrared light is absorbed by hemoglobin in the extracerebral and cerebral tissues. Therefore, OT signals are influenced by both scalp and cerebral hemodynamics. However, previous studies have shown that changes in cerebral hemodynamics and oxygenation due to manipulation of breathing gas can be detected in settings similar to those in the present study.1718.–19 In order to measure changes in concentrations of hemoglobin in the area of the MCA, probes were symmetrically set on the subject’s scalp to cover the bilateral fronto-temporal areas. Measurement was performed with a response rate of 10 Hz.
An oxygen face mask (OX-135, Atom Medical Corporation, Tokyo, Japan) was placed on the subject as shown in Fig. 1, and peripheral oxygen saturation () was measured at the tip of the subject’s index finger using a monitor (OLV 3100, Nihon Kohden Corporation, Tokyo, Japan). Air was delivered through the oxygen face mask at a flow rate of , and OT measurement was started. After OT data were stabilized, the air was switched to oxygen at the same flow rate for a period of 2 min, and then it was switched back to air at the same flow rate. The changes in showed a trapezoidal pattern with this manipulation. Changes in HbOxy associated with oxygen inhalation in the cerebral tissues were measured using OT. The subjects continued breathing in a free and relaxed manner during the measurement in order to minimize the influence of physiological factors that could cause changes in cerebral blood volume (CBV).20
OT Data Analysis
The time course data of HbOxy were analyzed using the following methods (Fig. 2). OT data were processed using MATLAB R2011b (MathWorks, Inc., Natick, MA).
Principal component analysis (PCA) was applied to the data and they were decomposed to principal components [Fig. 2(a)]. PCA has been widely used in data analysis to reduce data dimensionality without discarding essential features of the measurement data. PCA is a multivariate technique that utilizes synthetic variables (principal components), and it has been used to identify and remove signal interference from extracerebral tissues.2122.–23 PCA was performed by creating a variance-covariance matrix of all measurement data and calculating its eigenvalue and eigenvector. Weights were the coefficients for each channel for synthesis of the principal component, and they reflect the degree of contribution of the data in each channel to the principal component data. Therefore, if the wave’s height was low and/or the peak time was delayed, the weight is low.
Cross-correlation of each principal component and was calculated. It was determined that the principal component with the highest correlation coefficient included the information of changes in HbOxy associated with oxygen inhalation, and this was analyzed in detail.
In order to evaluate the transmission of the systemic changes to the cerebral tissues, weights of the principal components were calculated in each channel. The contour map (topogram) was drawn using the weights positioned at each measuring point. In the topogram, the values between each channel were calculated by spline interpolation [Fig. 2(b)].
Evaluation of Cerebral Ischemia
The laterality of OT was determined comparing the channel inspections in the symmetrical position on the topogram. The side with a lower weight was defined as the ischemic side. Statistical significance was determined using the Mann-Whitney U test as a nonparametric technique. Because not all the PCA weights had normal distributions, blood flows on the ischemic side and the contralateral side were considered unpaired groups. Differences with were considered significant.
OT findings were compared to those of SPECT, and the possibility of identification of cerebral ischemia was investigated. Moreover, in patients who underwent carotid artery stenting (CAS), the findings of SPECT and OT were compared before and after CAS. Detection of improvement of CBF associated with CAS was also evaluated.
In normal volunteers, changes in exhibited a trapezoidal pattern associated with oxygen inhalation, and the changes in HbOxy measured using OT showed a similar time course [Fig. 3(a)]. In the case shown in Fig. 3, there was no significant laterality of the weights of PCA (, Mann-Whitney U-test), and this finding was identified in the topogram [Fig. 3(b)]. These findings were observed in 25 (83%) of the normal volunteers.
On the other hand, with cerebral ischemia, the increase in HbOxy was delayed and/or lowered, and the PCA weights in the ischemic side were lower compared to the normal side. Therefore, the side with cerebral ischemia could be detected. Results in cases of cerebral ischemia are shown in Table 1. The ischemic side defined by SPECT and OT are shown in Table 1 as well. In 28 (85%) of the cerebral ischemia cases, the findings of OT agreed with those of SPECT. Moreover, in three patients who underwent CAS, OT findings were compared to those of SPECT and changes in CBF associated with CAS were compared. In the findings of SPECT for post-CAS patients, improvement in CBF was observed. On the other hand, in the findings of OT after CAS, increases in the PCA weights in the ischemic regions were observed. Laterality of the PCA weights was not observed after CAS. Over the time period of the measurement, HbTotal decreased slightly and HbR exhibited a wave pattern almost reverse to HbOxy.
Clinical features of cases with cerebral ischemia.
|Case||Atge||Sex||Vascular lesion||Ischemic side|
|1–1||73||M||L ICA stenosis||L||L|
|1–2||73||M||L ICA stenosis (post-CAS)||N||N|
|2||68||M||L MCA occlusion||L||L|
|3||65||M||L ICA stenosis||L||L|
|4–1||78||M||R ICA stenosis||R||R|
|4–2||78||M||R ICA stenosis (post-CAS)||N||N|
|5||70||M||L ICA stenosis||L||L|
|6||78||M||L ICA stenosis||L||L|
|7||67||M||L ICA stenosis||L||L|
|8–1||66||M||R ICA stenosis||R||R|
|8–2||66||M||R ICA stenosis (post-CAS)||N||N|
|9||58||M||R ICA stenosis||R||L|
|10||69||F||L MCA stenosis||L||L|
|11||69||F||R MCA occlusion||R||R|
|12||70||M||R ICA stenosis||R||R|
|13||61||M||R ICA occlusion||R||L|
|14||68||F||R MCA stenosis||R||L|
|15||78||M||R ICA occlusion||R||R|
|16||69||M||R ICA stenosis||R||L|
|17||78||M||L ICA stenosis||L||L|
|18||77||M||L ICA stenosis||L||L|
|19||61||M||L ICA stenosis||L||L|
|20||78||M||L ICA stenosis||L||L|
|21||67||M||L ICA stenosis||L||L|
|22||75||M||R ICA stenosis||R||R|
|23||67||M||L ICA stenosis||L||L|
|24||58||M||R ICA stenosis||R||R|
|25||60||M||R ICA occlusion||R||R|
|26||71||M||L ICA stenosis||L||L|
|27||61||M||R ICA occlusion||R||R|
|28||76||M||R ICA stenosis||R||R|
|29||60||M||L MCA stenosis||L||L|
|30||62||M||L ICA stenosis||L||L|
|31||65||F||L ICA stenosis||L||R|
|32||71||M||R ICA stenosis||R||R|
|33||78||M||R ICA stenosis||R||R|
R: right, L: left, N: no laterality
MCA: middle cerebral artery; ICA: internal carotid artery
SPECT: single-photon-emission computed tomography
OT: optical topography
The columns named SPECT and OT indicate the side with cerebral ischemia in each modality.
Typical Case of Cerebral Ischemia
A 73-year-old male patient had severe stenosis of his left cervical ICA that was observed angiographically [Fig. 4(a)]. SPECT demonstrated a decrease in CBF in the left cerebral hemisphere [Fig. 4(b)]. CAS was performed on the left cervical ICA [Fig. 4(c)]. Improvement of CBF was observed in SPECT after CAS [Fig. 4(d)]. On the other hand, in the left cerebral hemisphere, the increase in HbOxy measured using OT with oxygen inhalation was lower than that of the contralateral side before CAS [Fig. 5(a), Table 1: Case 1-1]. However, the laterality of the changes in HbOxy disappeared after CAS because the changes in HbOxy were associated with CAS [Fig. 5(c), Table 1: Case 1-2].
In this case, OT measurement data were analyzed statistically by PCA, and the correlation coefficients of each principal component and were calculated. Before CAS, the correlation coefficient of the first principal component and was the highest (), so the first principal component was selected as the component that included information associated with oxygen inhalation. Weights from both sides were compared in this principal component, and the weights in the left side were significantly decreased compared to the weights in the contralateral side (, Mann-Whitney U-test). Topogram findings agreed with the regions of cerebral ischemia detected in SPECT [Fig. 5(b)].
After CAS, the correlation coefficient of the first principal component and was found to be the highest (), and this principal component was selected as the component that included information associated with oxygen inhalation. Weights of this principal component were calculated, and the weights from both sides were compared. There was no significant laterality of the PCA weights (, Mann-Whitney U-test). Topogram findings agreed with the regions with improvement of CBF in SPECT after CAS [Fig. 5(d)]. Averages of the PCA weights in normal volunteers and ischemic patients are shown in Fig. 6.
In normal volunteers, the changes in HbOxy exhibited a trapezoidal pattern according to oxygen inhalation, and there was no laterality of the PCA weights based on the wave pattern in 83% of cases. Therefore, this finding was regarded as typical of OT in normal tissues. The laterality of findings was also investigated. In all cases with no laterality of the PCA weights, HbTotal was slightly decreased and did not fluctuate. However, in cases with laterality, HbTotal was unstable. This instability was thought to be because cerebral hemodynamic fluctuation of a systemic origin was insufficiently removed from the cases with laterality.
On the other hand, in the ischemic cases, increases in HbOxy were lower in the ischemic regions compared to the normal regions. The PCA weights based on the wave pattern were lower in the ischemic regions in the topogram. These findings agreed with those of SPECT in 85% of ischemic cases. The reason for this was considered to be that transportation and diffusion of oxygen were attenuated in the ischemic regions by stenosis or occlusion of the ICA or MCA. In ischemic cases with no laterality in the PCA weights, CBF decreased slightly compared to the contralateral side in the findings of SPECT. It was thought that detection of mild ischemia was difficult with the present method. Cerebral oxygenation in regions with mild ischemia was the same as in normal regions. This is because the decrease in CBF leads to an increase in oxygen extraction fraction in order to preserve oxygen metabolism.24,25
Moreover, in the ischemic cases, findings of SPECT before and after CAS were compared. In the regions where CBF was lower than the normal regions before CAS, increases in CBF were observed after CAS and laterality was not observed. In these cases, changes in HbOxy measured using OT increased after CAS in regions where they had been lower than the normal regions before CAS. PCA weights increased after CAS, and this result was also observed in the topogram that was drawn using the weights. Regions where the weights increased agreed with regions that showed improvement in CBF as detected by SPECT.
It is important to note that there is unavoidable influence that comes from the CBV of the scalp because the concentration of hemoglobin is measured through the scalp. It has been reported that oxygen inhalation raises skin blood flow moderately, but that the change in flow is not significant.17 Although PCA could not entirely remove the extracerebral contribution from OT signals collected with several channels, it was useful for reducing signal interference from extracerebral tissues.
In the present cases of cerebral ischemia, improvement of blood flow associated with CAS was only in the ICA, and blood flow in the external carotid artery (ECA) did not change. Therefore, differences in the findings of OT before and after CAS must be attributed mainly to changes in CBF from the ICA, with only a small influence due to blood flow from the ECA.
OT has mainly been used to visualize brain functions in which changes in CBV are measured during a physiological task. On the other hand, in the present method, HbOxy in cerebral tissues is measured during oxygen inhalation. It has been reported that cerebral perfusion can be measured by OT using an intravenous bolus of indocyanine green.26,27 The present method is similar to that technique and can be used to detect a difference in the efficiency of transport and diffusion of oxygen to the cerebral tissues.
M. Komiyamaet al., “Prospective analysis of complications of catheter cerebral angiography in the digital subtraction angiography and magnetic resonance era,” Neurol. Med. Chir. 38(9), 534–540 (1998).0470-8105http://dx.doi.org/10.2176/nmc.38.534Google Scholar
M. Wintermarket al., “Accuracy of dynamic perfusion CT with deconvolution in detecting acute hemispheric stroke,” AJNR Am. J. Neuroradiol. 26(1), 104–112 (2005).0195-6108Google Scholar
B. Wuet al., “Collateral circulation imaging: MR perfusion territory arterial spin-labeling at 3T,” AJNR Am. J. Neuroradiol. 29(10), 1855–1860 (2008).0195-6108http://dx.doi.org/10.3174/ajnr.A1259Google Scholar
A. M. Demchuket al., “Specific transcranial doppler flow findings related to the presence and site of arterial occlusion,” Stroke 31(1), 140–146 (2000).SJCCA70039-2499http://dx.doi.org/10.1161/01.STR.31.1.140Google Scholar
F. F. Jöbsis, “Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters,” Science 198(4323), 1264–1267 (1977).SCIEAS0036-8075http://dx.doi.org/10.1126/science.929199Google Scholar
E. Watanabeet al., “Non-invasive functional mapping with multi-channel near infra-red spectroscopic topography in humans,” Neurosci. Lett. 205(1), 41–44 (1996).NELED50304-3940http://dx.doi.org/10.1016/0304-3940(96)12376-4Google Scholar
K. Isobeet al., “Functional imaging of the brain in sedated newborn infants using near-infrared topography during passive knee movement,” Neurosci. Lett. 299(3), 221–224 (2001).NELED50304-3940http://dx.doi.org/10.1016/S0304-3940(01)01518-XGoogle Scholar
P. H. KohC. E. ElwellD. T. Delpy, “Development of a dynamic test phantom for optical topography,” Adv. Exp. Med. Biol. 645, 141–146 (2009).AEMBAP0065-2598http://dx.doi.org/10.1007/978-0-387-85998-9Google Scholar
E. Watanabeet al., “Noninvasive cerebral blood volume measurement during seizures using multichannel near-infrared spectroscopic topography,” J. Biomed. Opt. 5(3), 287–290 (2000).JBOPFO1083-3668http://dx.doi.org/10.1117/1.429998Google Scholar
R. P. Kannanet al., “Non-invasive assessment of language lateralization by transcranial near-infrared optical topography and functional MRI,” Hum. Brain Mapp. 16(3), 183–189 (2002).HBRME71065-9471http://dx.doi.org/10.1002/(ISSN)1097-0193Google Scholar
K. MatsuoT. KatoM. FukudaN. Kato, “Alteration of hemoglobin oxygenation in the frontal region in elderly depressed patients as measured by near-infrared spectroscopy,” J. Neuropsychiatry Clin. Neurosci. 12(4), 465–471 (2000).Google Scholar
A. Casatiet al., “New technology for noninvasive brain monitoring: continuous cerebral oximetry,” Minerva Anesthesiol. 72(7–8), 605–625 (2006).0375-9393Google Scholar
S. Nakajimaet al., “Cerebral vasomotor responsiveness during 100% oxygen inhalation in cerebral ischemia,” Arch. Nerol. 40(5), 271–276 (1983).0003-9942http://dx.doi.org/10.1001/archneur.1983.04050050039004Google Scholar
P. Smilewskiet al., “Clinical evaluation of near-infrared spectroscopy for testing cerebrovascular reactivity in patients with carotid artery disease,” Stroke 28(2), 331–338 (1997).SJCCA70039-2499http://dx.doi.org/10.1161/01.STR.28.2.331Google Scholar
C. Terborget al., “Noninvasive monitoring of cerebral oxygenation during vasomotor reactivity tests by a new near-infrared spectroscopy device,” Cerebrovasc. Dis. 16(1), 36–41 (2003).1015-9770http://dx.doi.org/10.1159/000070113Google Scholar
C. E. Elwellet al., “Quantification of adult cerebral hemodynamics by near-infrared spectroscopy,” J. Appl. Physiol. 77(6), 2753–2760 (1994).JAPYAA0021-8987Google Scholar
T. J. Germonet al., “Sensitivity of near infrared spectroscopy to cerebral and extra-cerebral oxygenation changes is determined by emitter-detector separation,” J. Clin. Monit. Comput. 14, 353–360 (1998).1387-1307Google Scholar
R. J. Cooperet al., “The utility of near-infrared spectroscopy in the regression of low-frequency physiological noise from functional magnetic resonance imaging data,” Neuroimage 59(4), 3128–3138 (2012).NEIMEF1053-8119http://dx.doi.org/10.1016/j.neuroimage.2011.11.028Google Scholar
J. VirtanenT. NoponenP. Meriläinen, “Comparison of principal and independent component analysis in removing extracerebral interference from near-infrared spectroscopy signals,” J. Biomed. Opt. 14(5), 054032 (2009).JBOPFO1083-3668http://dx.doi.org/10.1117/1.3253323Google Scholar
Y. Zhanget al., “Eigenvector-based spatial filtering for reduction of physiological interference in diffuse optical imaging,” J. Biomed. Opt. 10(1), 11014 (2005).JBOPFO1083-3668http://dx.doi.org/10.1117/1.1852552Google Scholar
A. CaprihanG. D. PearlsonV. D. Calhoun, “Application of principal component analysis to distinguish patients with schizophrenia from healthy controls based on fractional anisotropy measurements,” Neuroimage 42(2), 675–682 (2008).NEIMEF1053-8119http://dx.doi.org/10.1016/j.neuroimage.2008.04.255Google Scholar
W. J. PowersR. L. Grubb Jr.M. E. Raichle, “Physiological responses to focal cerebral ischemia in humans,” Ann. Neurol. 16(5), 546–552 (1984).ANNED30364-5134http://dx.doi.org/10.1002/(ISSN)1531-8249Google Scholar
C. P. Derdeynet al., “Variability of cerebral blood volume and oxygen extraction: stages of cerebral haemodynamic impairment revisited,” Brain 125(3), 595–607 (2002).BRAIAK0006-8950http://dx.doi.org/10.1093/brain/awf047Google Scholar
W. M. Kuebleret al., “Noninvasive measurement of regional cerebral blood flow by near-infrared spectroscopy and indocyanine green,” J. Cereb. Blood Flow Metab. 18(4), 445–456 (1998).0271-678Xhttp://dx.doi.org/10.1097/00004647-199804000-00013Google Scholar
C. Terborget al., “Bedside assessment of cerebral perfusion reductions in patients with acute ischemic stroke by near-infrared spectroscopy and indocyanine green,” J. Neurol. Neurosurg. Psychiatry 75(1), 38–42 (2004).JNNPAU0022-3050Google Scholar