J Neurosurg 76:315-318, 1992
Intracerebral penetration of infrared light Technical note PATRICK W. McCORMICK, M.D., MELVILLE STEWART, M.S., GARY LEWIS, E.E., MANUEL DUJOVNY, M.D., AND JAMES I. AUSMAN, M.D., PH.D. Henry Ford Neurosurgical Institute, Henry Ford Hospital, Detroit, Michigan ~," Near infrared transmission spectroscopy of the human cerebrum may allow noninvasive evaluation of cerebral hemoglobin saturation in humans. The emerging spectroscopyconfiguration for this application is a side-by-sidesource-receiverconstruct. The ability of this spectroscopyparadigm to detect changes in intracerebral attenuation by selectiveinjection of the infrared tracer indocyanine green into the internal and external carotid arteries during endarterectomy is evaluated in five adult patients. In all five, simultaneous two-channel infrared transmission spectroscopy over the ipsilateral hemisphere documented tracer bolus transit with a signal-to-noise ratio greater than 100:1. In addition, the two channels could be configuredto achieve depth resolution of the collected spectra. KEY WORDS
~
indocyanine green
infrared spectroscopy 9 oximetry 9 intravascular tracers
HE use of infrared light for noninvasive monitoring of important physiological parameters, such as arterial hemoglobin oxygen saturation, has become ubiquitous in clinical practice in the form of pulse oximetry.28 These quantitative on-line measurements are made by analyzing the wavelength-dependent attenuation of infrared light as it passes through the tissue of interest, s'25 Such a paradigm is commonly referred to as in vivo optical spectroscopy. Interest has focused over the last 10 years on applying this technology to the noninvasive measurement of cerebral oxyhemoglobin concentration and cerebral hemoglobin oxygen saturation. ~4There are several reports of noninvasive, qualitative, and quantitative measurement of these parameters in laboratory animals, infants, and adults.2-4'6-12'14'ls,Iq,2z27,29-31 An issue fundamental to the clinical application of in vivo optical spectroscopy for cerebral measurement is whether infrared light in the range of 650 to 1100 nm can penetrate through the adult human scalp and skull, undergo attenuation specific to the intracerebral compartment, and return to the scalp for detection. Chance, el al.,4 and Delpy, et al.,7 pioneers in this area, have independently demonstrated the theoretical fea-
T
J. Neurosurg. / Volume 76/February, 1992
9 tissue spectroscopy 9
sibility of this based on the decay kinetics of pulsed, coherent, infrared light as it propagates through human cerebral tissue or optical models of human cranial tissues. The successful paradigm they have identified is called "diffuse transmission spectroscopy" and involves propagating light intracranially from a point source and collecting it ipsilateral to the point source.4'7 Diffuse transmission spectroscopy is performed by placing an infrared light source and receiver in relatively close proximity on the surface of the scalp. ~7.18~23Interestingly, this configuration of light source and receiver does not result in light traveling along the shortest possible path between the two; rather, the highest probability photon path penetrates into the surrounding tissue. An example of this tissue penetration is laser Doppler flowimetry. Here, an infrared light source and receiver are placed 1 to 2 mm apart, yet a strong signal from the tissue microvasculature is detected.24 In a side-by-side configuration, the depth of penetration is a function of the source-receiver separation distance; the greater this distance is, the deeper the penetration of the mean tissue photon path of the tissue being sampled, j'2~ It remains to be demonstrated whether the configuration of a single point source and one or more ipsilateral re315
P. W. McCormick, et al.
FIG. 1. The interface of infrared source and receiverfibers in the cranium of an adult subject. Infrared light propagates through the scalp, skull, and cerebral parenchyma and returns to receiver fibers via the high probability photon paths demonstrated (arrow).lnset: The interfacecontains a singlesource bundle and two receiver bundles. This configuration results in transmission of light over two different photon paths. The closer receiver is effected by attenuation in the superficial tissue (small arrow) and the further receiver by attenuation in both the superficial and deep tissue (large arrow).
ceivers is capable of detecting intracranial infrared light attenuation through the intact scalp and skull of adult humans. Clinical Material and Methods Five adults undergoing routine carotid endarterectomy participated in this study under a protocol approved by the Henry Ford Hospital Human Rights in Experimentation Committee. In these patients, a bolus (1.0 mg in 1.0 cc normal saline) of the infrared attenuating tracer indoeyanine green* was placed into the exposed internal carotid artery (ICA) by cannulation with a No. 22 needle. The external carotid artery was temporarily occluded to prevent reflux of tracer into this vessel. Diffuse transmission spectroscopy was carried out over the ipsilateral frontal bone 4.0 cm anterior to the coronal suture and 3.0 cm lateral to the sagittal suture using a noncoherent, nonpulsed, 803-nm light delivered to a point source via silicon oxide fiberoptic bundles.t A two-channel receiver tuned to 803 nm was used to collect simultaneous infrared light signals 1.0 cm and 2.7 em from the point source (Fig. 1). Analysis of infrared light propagation in human tissues suggests that the mean photon path to the 1.0-cm receiver should not penetrate below the skull ("superficial" channel), 2~The 2.7-cm receiver was chosen because the mean photon path from source to receiver should penetrate 1.2 to 2.0 cm into the brain ("deep" channel). After the indocyanine green bolus was intro* Indoeyanine green supplied by Becton Dickinson Co., Baltimore, Maryland. ~"INVOS 2910 fiberoptic bundles manufactured by Somanetics, Troy, Michigan.
316
FIG. 2. Graph showingattenuation at 803 nm versus time following introduction of a bolus of indoeyanine green into the internal carotid artery. Results at 8, l 5, 40, and 60 seconds are shown and each represents the average findings in five subjects. The upper line (squares) representsa rapid increase in attenuation followedby a slightlyslower return to baseline in the 2.7-emsource-receiverseparationdistance channel. The lower line (circles)represents the 1.0-cm source-receiverseparation channel with little change in attenuation.
duced into the ICA, a series of transmission spectroscopic measurements were made for a period of 1 minute at a rate of 29/sec. The transit of infrared absorber through the intracranial vascular compartment was detected as the change in optical density at 803 nm over time. The measurement was then repeated allowing the tracer to flow only into the external carotid artery. Results The indocyanine green bolus infusion did not cause any significant change in vital signs, and no patient suffered associated surgical morbidity or mortality. Bolus infusion into the ICA did not significantly alter the optical density recorded in the channel with a 1.0-cm source-receiver separation distance (Fig. 2). This is to be expected because the superficial extracranial tissue sampled by this source-receiver configuration is not vascularized by the ICA. In contrast, major changes in the optical density were recorded in the channel with a 2.7-cm source-receiver separation distance following ICA tracer infusion (Fig. 2). A significant portion of the attenuating events experienced by photons reaching this receiver is predicted to take place intracranially where tracer-induced attenuation should be large. The bolus could be detected washing into and out of the cerebral blood vessels in the field included by this deeper sampling channel. This activity curve is entirely compatible with a nondiffusible tracer, such as indocyanine green, transiting the cerebrovascular bed and is a measure of regional mean cerebral transit time through the sampled vessels. The change in optical density in the two channels with an ICA bolus of tracer was very reproducible in all five subjects (Fig. 2).
J. Neurosurg. / Volume 76 / February, 1992
Intracerebral penetration of infrared light
FIG. 3. Graph showingattenuation at 803 nm versus time followingintroduction of a bolus of indocyanine green in the external carotid artery. Results at 6, 12, 40, and 60 seconds are shown and each represents the average findings in five subjects. The upper line (squares) representsthe rapid increase in attenuation followedby a return to baseline in the 2.7-cm source-receiver separation distance channel. The lower line (circles) represents the 1.0-era source-receiver separation channel with an almost identical response to the channel above.
When the tracer was allowed to flow exclusively into the external carotid artery, both receivers generated activity curves similar to those in the 2.7-cm sourcereceiver separation channel during ICA infusion (Fig. 3). Here the tracer is in extracranial vascular beds, and therefore in the field of both receivers. The activity curves demonstrate the "wash in" and "wash out" of indocyanine green in the extracranial vascular beds. The signal-to-noise ratio for detection of indocyanine green in the cerebrovasculature in the 2.7-cm sourcereceiver separation channel is greater than 100:1. Signal was defined as the maximum increase in optical density associated with indocyanine green infusion, and noise was defined as the peak-to-peak baseline optical density oscillation. The apparent noise was due primarily to systolic and diastolic fluctuations in blood concentration. Discussion
The concept of using in vivo optical spectroscopy to make noninvasive on-line measurements of oxygen saturation in the cerebrovascular compartment is appealing. The introduction of diffuse transmission spectroscopy paradigms has made this appear theoretically feasible. It has not been demonstrated that infrared light is capable of substantial intracranial penetration or that light-attenuating events in the intracranial compartment can be successfully detected by surface recording. This experimental protocol was designed to determine whether the ipsilateral configuration of a near infrared light source and receiver allowed noninvasive detection of changes in intracranial infrared light attenuation. The dual-channel continuous wavelength specJ. Neurosurg. / Volume 76/February, 1992
troscopy apparatus recorded the changes in intracranial light attenuation through the intact scalp and skull. These changes were induced by selectively altering the optical density of the cerebrovasculature with an intravascular bolus of infrared attenuating tracer. The data demonstrate the ability of infrared light near the 803-nm waveband to penetrate living human scalp and skull, undergo attenuation in the cerebrovascular compartment, and manifest this attenuation as an altered transmission intensity. It is also evident from the data that the peculiarities of light propagation through human tissue allow a certain degree of depth resolution of collected spectra. Superficial (extracranial) attenuation can be differentiated from deep (intracranial) attenuation. The theoretical explanation for this is emerging from sophisticated computer modeling studies. 2~ Bonnet, et al./have predicted the dependence of tissue penetration on source-receiver configuration and spatial separation. This is the first report demonstrating depth resolution of cerebral infrared transmission spectra in humans. In this protocol, infrared attenuation from superficial and deep tissues was clearly differentiated in two channels. The potential for multiple channels is clearly present. It is important to note that infrared light transmission through the living human scalp, calvaria, and brain is similar to that predicted by theoretical analysis and modeling paradigms. The practical demonstration of remote depth-resolved detection has the relatively immediate potential application of noninvasive continuous monitoring of cerebral oxygen saturation. ~3'~5'~7'~8,26,32 Indocyanine green signals are detected with an excellent signal-to-noise ratio and generate an analog function of bolus transit through the cerebrovasculature. The detecting apparatus is regional, and mapping of cerebral mean transit time is feasible with this technique. ]~ Resolution of superficial (extracranial) and deep (cerebral) transit times should be possible with multiple receiver paradigms. References
1. Bonner RF, Nossal R, Havlin S, et al: Model for photon migration in turbid biological media, d Opt Soc Am (A) 4:423-432, 1987 2. Brazy JE: Effectsof crying on cerebral blood volume and cytochrome aa~. J Pefliao"112:457-461, 1988 3. Brazy JE, Lewis DV, Mitnick MH, et al: Noninvasive monitoring of cerebral oxygenation in preterm infants: preliminary observations. Pediatrics 75:217-225, 1985 4. Chance B, Leigh JS, Miyake H, et al: Comparison of time-resolved and -unresolved measurements of deoxyhemoglobin in brain. Proc Nati Acad Sci USA 85: 4971--4975, 1988 5. Cohen A, Wadsworth N: A light emitting diode skin reflectanceoximeter. Med Biol Eng 10:385-391, 1972 6. Cope M, Delpy DT, Reynolds EO, el al: Methods of quantitating cerebral near infrared spectroscopydata. Adv Exp Med Biol 222:183-189, 1988 7. Delpy DT, Cope M, van der Zee P, et al: Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 33:1433-1442, 1988 317
P. W. McCormick, etal. 8. Delpy DT, Cope MC, Cady EB, et al: Cerebral monitoring in newborn infants by magnetic resonance and near infrared spectroscopy. Scand J Clin Lab Invest (Suppl) 188: 9-17, 1987 9. Ferrari M, De Marchis C, Giannini I, el al: Cerebral blood volume and hemoglobin oxygen saturation monitoring in neonatal brain by near IR spectroscopy. Aflv Exp Med Biol 200:203-211, 1986 10. Giannini I, Ferrari M, CarpiA, etal: Rat brain monitoring by near-infrared spectroscopy: an assessment of possible clinical significance. Physiol Chem Phys 14:295-305, 1982 11. Hazeki O, Seiyama A, Tamura M: Near-infrared spectrophotometric monitoring of haemoglobin and eytochrome a, a3 in situ. Adv Exp Mefl Bid 215:283-289, 1987 12. Hazeki O, Tamura M: Quantitative analysis of hemoglobin oxygenation state of rat brain in situ by near-infrared spectrophotometry. J Appl Physio164:796-802, 1988 13. Hoffmann J, LObbers DW: Estimation of concentration ratios and the redox states of the cytochromes from noisy reflection spectra using multicomponent analysis methods. Adv Exp Med Binl 200:119-124, 1986 14. Jrbsis FF: Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 198:1264-1267, 1977 15. Mayevsky A, Nioka S, Chance B: Fiber optic surface fluoremetry/reflectometryand 3 I-p-NMR for monitoring the intracellular energy state in vivo. Adv Exp Med Biol 222:365-374, 1988 16, McCormick PW, Stewart M, Goetting M, el al: The use of nondiffusible optically active dye tracer for noninvasive quantification of cerebral blood flow. Stroke 20:134, 1989 (Abstract) 17. McCormick PW, Stewart M, Goetting MG, etal: Noninvasive cerebral optical spectroscopy for monitoring cerebral oxygen delivery and hemodynamics. Crit Care Med 19:89-97, 1991 18. McCormick PW, Stewart M, Goetting MG, etal: Regional cerebrovascular oxygen saturation measured by optical spectroscopy in humans. Stroke 22:596-602, 1991 19. McCormick PW, Stewart M, Ray P, el al: Measurement of regional cerebrovascular haemoglobin oxygen saturation in cats using optical spectroscopy. Neurol Res 13: 65-70, 1991 20. Nossal R, Bonner RF, Weiss GH: The influence of path length on remote optical sensing of properties of biological tissue. Appl Optics 28:2238-2244, 1989 21. Patterson MS, Chance B, Wilson BC: Time resolved
318
reflectance and transmittance for the noninvasive measurement of tissue optical properties. Appl Optics 28: 2331-2336, 1989 22. Seeds JW, Cefalo RC, Proctor HJ, etaI: The relationship of intracranial infrared light absorbance to fetal oxygenation. I: methodology. Am J Obstet Gyneeol 149: 679-684, 1984 23. Smith DS, Levy W, Maris M, etal: Reperfusion hyperoxia in brain after circulatory arrest in humans. Anesthesiology 73:12-19, 1990 24. Stern MD: In vivo evaluation of microcirculation by coherent light scattering. Nature 254:56--58, 1975 25. Takatani S, Cheung PW, Ernst EA: A noninvasive tissue reflectance oximeter. An instrument for measurement of tissue hemoglobin oxygen saturation in vivo. Ann Biochem Eng 8:1-15, 1980 26. Tamura T, Hazeki O, Nioka S, et al: In vivo study of tissue oxygen metabolism using optical and nuclear magnetic resonance spectroscopy. Annu Rev Physiol 51: 813-834, 1989 27. Tamura T, Hazeki O, Takada M, etal: Abserbance profile of red blood cell suspension in vitro and in situ. Adv Exp Med Biol 222:211-217, 1988 28. Taylor MB, Whitwam JG: The current status of pulse oximetry. Clinical value of continuous noninvasive oxygen saturation monitoring. Anesthesia 41:943-949, 1987 29. Wiernsperger N, Sylvia AL, J6bsis FF: Incomplete transient ischemia: a nondestructive evaluation of in vivo cerebral metabolism and hemodynamics in rat brain. Stroke 12:864-868, 1981 30. Wray S, Cope M, Delpy DT, et al: Characterization of the near infrared absorption spectra of cytochrome aa 3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochim Biophys Acta 933:184-192, 1988 31. Wyatt JS, Cope M, Delpy DT, et at: Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 2:1063-1066, 1986 32. Yamashita Y, Suzuki S, Miyaki S, etal: The neonate brain and breast imaging using NIR transillumination, in Chance B (ed): Photon Migration in Tissues. New York: Academic Press, 1990, pp 55-67 Manuscript received April 2, 1991. Accepted in final form July 16, 1991. Address reprint requests to." Patrick W. McCormick, M.D., 2213 Cherry Street, Suite 311, Toledo, Ohio 43608.
J. Neurosurg. / Volume 76 /February, 1992
Intracerebral penetration of infrared light Technical note PATRICK W. McCORMICK, M.D., MELVILLE STEWART, M.S., GARY LEWIS, E.E., MANUEL DUJOVNY, M.D., AND JAMES I. AUSMAN, M.D., PH.D. Henry Ford Neurosurgical Institute, Henry Ford Hospital, Detroit, Michigan ~," Near infrared transmission spectroscopy of the human cerebrum may allow noninvasive evaluation of cerebral hemoglobin saturation in humans. The emerging spectroscopyconfiguration for this application is a side-by-sidesource-receiverconstruct. The ability of this spectroscopyparadigm to detect changes in intracerebral attenuation by selectiveinjection of the infrared tracer indocyanine green into the internal and external carotid arteries during endarterectomy is evaluated in five adult patients. In all five, simultaneous two-channel infrared transmission spectroscopy over the ipsilateral hemisphere documented tracer bolus transit with a signal-to-noise ratio greater than 100:1. In addition, the two channels could be configuredto achieve depth resolution of the collected spectra. KEY WORDS
~
indocyanine green
infrared spectroscopy 9 oximetry 9 intravascular tracers
HE use of infrared light for noninvasive monitoring of important physiological parameters, such as arterial hemoglobin oxygen saturation, has become ubiquitous in clinical practice in the form of pulse oximetry.28 These quantitative on-line measurements are made by analyzing the wavelength-dependent attenuation of infrared light as it passes through the tissue of interest, s'25 Such a paradigm is commonly referred to as in vivo optical spectroscopy. Interest has focused over the last 10 years on applying this technology to the noninvasive measurement of cerebral oxyhemoglobin concentration and cerebral hemoglobin oxygen saturation. ~4There are several reports of noninvasive, qualitative, and quantitative measurement of these parameters in laboratory animals, infants, and adults.2-4'6-12'14'ls,Iq,2z27,29-31 An issue fundamental to the clinical application of in vivo optical spectroscopy for cerebral measurement is whether infrared light in the range of 650 to 1100 nm can penetrate through the adult human scalp and skull, undergo attenuation specific to the intracerebral compartment, and return to the scalp for detection. Chance, el al.,4 and Delpy, et al.,7 pioneers in this area, have independently demonstrated the theoretical fea-
T
J. Neurosurg. / Volume 76/February, 1992
9 tissue spectroscopy 9
sibility of this based on the decay kinetics of pulsed, coherent, infrared light as it propagates through human cerebral tissue or optical models of human cranial tissues. The successful paradigm they have identified is called "diffuse transmission spectroscopy" and involves propagating light intracranially from a point source and collecting it ipsilateral to the point source.4'7 Diffuse transmission spectroscopy is performed by placing an infrared light source and receiver in relatively close proximity on the surface of the scalp. ~7.18~23Interestingly, this configuration of light source and receiver does not result in light traveling along the shortest possible path between the two; rather, the highest probability photon path penetrates into the surrounding tissue. An example of this tissue penetration is laser Doppler flowimetry. Here, an infrared light source and receiver are placed 1 to 2 mm apart, yet a strong signal from the tissue microvasculature is detected.24 In a side-by-side configuration, the depth of penetration is a function of the source-receiver separation distance; the greater this distance is, the deeper the penetration of the mean tissue photon path of the tissue being sampled, j'2~ It remains to be demonstrated whether the configuration of a single point source and one or more ipsilateral re315
P. W. McCormick, et al.
FIG. 1. The interface of infrared source and receiverfibers in the cranium of an adult subject. Infrared light propagates through the scalp, skull, and cerebral parenchyma and returns to receiver fibers via the high probability photon paths demonstrated (arrow).lnset: The interfacecontains a singlesource bundle and two receiver bundles. This configuration results in transmission of light over two different photon paths. The closer receiver is effected by attenuation in the superficial tissue (small arrow) and the further receiver by attenuation in both the superficial and deep tissue (large arrow).
ceivers is capable of detecting intracranial infrared light attenuation through the intact scalp and skull of adult humans. Clinical Material and Methods Five adults undergoing routine carotid endarterectomy participated in this study under a protocol approved by the Henry Ford Hospital Human Rights in Experimentation Committee. In these patients, a bolus (1.0 mg in 1.0 cc normal saline) of the infrared attenuating tracer indoeyanine green* was placed into the exposed internal carotid artery (ICA) by cannulation with a No. 22 needle. The external carotid artery was temporarily occluded to prevent reflux of tracer into this vessel. Diffuse transmission spectroscopy was carried out over the ipsilateral frontal bone 4.0 cm anterior to the coronal suture and 3.0 cm lateral to the sagittal suture using a noncoherent, nonpulsed, 803-nm light delivered to a point source via silicon oxide fiberoptic bundles.t A two-channel receiver tuned to 803 nm was used to collect simultaneous infrared light signals 1.0 cm and 2.7 em from the point source (Fig. 1). Analysis of infrared light propagation in human tissues suggests that the mean photon path to the 1.0-cm receiver should not penetrate below the skull ("superficial" channel), 2~The 2.7-cm receiver was chosen because the mean photon path from source to receiver should penetrate 1.2 to 2.0 cm into the brain ("deep" channel). After the indocyanine green bolus was intro* Indoeyanine green supplied by Becton Dickinson Co., Baltimore, Maryland. ~"INVOS 2910 fiberoptic bundles manufactured by Somanetics, Troy, Michigan.
316
FIG. 2. Graph showingattenuation at 803 nm versus time following introduction of a bolus of indoeyanine green into the internal carotid artery. Results at 8, l 5, 40, and 60 seconds are shown and each represents the average findings in five subjects. The upper line (squares) representsa rapid increase in attenuation followedby a slightlyslower return to baseline in the 2.7-emsource-receiverseparationdistance channel. The lower line (circles)represents the 1.0-cm source-receiverseparation channel with little change in attenuation.
duced into the ICA, a series of transmission spectroscopic measurements were made for a period of 1 minute at a rate of 29/sec. The transit of infrared absorber through the intracranial vascular compartment was detected as the change in optical density at 803 nm over time. The measurement was then repeated allowing the tracer to flow only into the external carotid artery. Results The indocyanine green bolus infusion did not cause any significant change in vital signs, and no patient suffered associated surgical morbidity or mortality. Bolus infusion into the ICA did not significantly alter the optical density recorded in the channel with a 1.0-cm source-receiver separation distance (Fig. 2). This is to be expected because the superficial extracranial tissue sampled by this source-receiver configuration is not vascularized by the ICA. In contrast, major changes in the optical density were recorded in the channel with a 2.7-cm source-receiver separation distance following ICA tracer infusion (Fig. 2). A significant portion of the attenuating events experienced by photons reaching this receiver is predicted to take place intracranially where tracer-induced attenuation should be large. The bolus could be detected washing into and out of the cerebral blood vessels in the field included by this deeper sampling channel. This activity curve is entirely compatible with a nondiffusible tracer, such as indocyanine green, transiting the cerebrovascular bed and is a measure of regional mean cerebral transit time through the sampled vessels. The change in optical density in the two channels with an ICA bolus of tracer was very reproducible in all five subjects (Fig. 2).
J. Neurosurg. / Volume 76 / February, 1992
Intracerebral penetration of infrared light
FIG. 3. Graph showingattenuation at 803 nm versus time followingintroduction of a bolus of indocyanine green in the external carotid artery. Results at 6, 12, 40, and 60 seconds are shown and each represents the average findings in five subjects. The upper line (squares) representsthe rapid increase in attenuation followedby a return to baseline in the 2.7-cm source-receiver separation distance channel. The lower line (circles) represents the 1.0-era source-receiver separation channel with an almost identical response to the channel above.
When the tracer was allowed to flow exclusively into the external carotid artery, both receivers generated activity curves similar to those in the 2.7-cm sourcereceiver separation channel during ICA infusion (Fig. 3). Here the tracer is in extracranial vascular beds, and therefore in the field of both receivers. The activity curves demonstrate the "wash in" and "wash out" of indocyanine green in the extracranial vascular beds. The signal-to-noise ratio for detection of indocyanine green in the cerebrovasculature in the 2.7-cm sourcereceiver separation channel is greater than 100:1. Signal was defined as the maximum increase in optical density associated with indocyanine green infusion, and noise was defined as the peak-to-peak baseline optical density oscillation. The apparent noise was due primarily to systolic and diastolic fluctuations in blood concentration. Discussion
The concept of using in vivo optical spectroscopy to make noninvasive on-line measurements of oxygen saturation in the cerebrovascular compartment is appealing. The introduction of diffuse transmission spectroscopy paradigms has made this appear theoretically feasible. It has not been demonstrated that infrared light is capable of substantial intracranial penetration or that light-attenuating events in the intracranial compartment can be successfully detected by surface recording. This experimental protocol was designed to determine whether the ipsilateral configuration of a near infrared light source and receiver allowed noninvasive detection of changes in intracranial infrared light attenuation. The dual-channel continuous wavelength specJ. Neurosurg. / Volume 76/February, 1992
troscopy apparatus recorded the changes in intracranial light attenuation through the intact scalp and skull. These changes were induced by selectively altering the optical density of the cerebrovasculature with an intravascular bolus of infrared attenuating tracer. The data demonstrate the ability of infrared light near the 803-nm waveband to penetrate living human scalp and skull, undergo attenuation in the cerebrovascular compartment, and manifest this attenuation as an altered transmission intensity. It is also evident from the data that the peculiarities of light propagation through human tissue allow a certain degree of depth resolution of collected spectra. Superficial (extracranial) attenuation can be differentiated from deep (intracranial) attenuation. The theoretical explanation for this is emerging from sophisticated computer modeling studies. 2~ Bonnet, et al./have predicted the dependence of tissue penetration on source-receiver configuration and spatial separation. This is the first report demonstrating depth resolution of cerebral infrared transmission spectra in humans. In this protocol, infrared attenuation from superficial and deep tissues was clearly differentiated in two channels. The potential for multiple channels is clearly present. It is important to note that infrared light transmission through the living human scalp, calvaria, and brain is similar to that predicted by theoretical analysis and modeling paradigms. The practical demonstration of remote depth-resolved detection has the relatively immediate potential application of noninvasive continuous monitoring of cerebral oxygen saturation. ~3'~5'~7'~8,26,32 Indocyanine green signals are detected with an excellent signal-to-noise ratio and generate an analog function of bolus transit through the cerebrovasculature. The detecting apparatus is regional, and mapping of cerebral mean transit time is feasible with this technique. ]~ Resolution of superficial (extracranial) and deep (cerebral) transit times should be possible with multiple receiver paradigms. References
1. Bonner RF, Nossal R, Havlin S, et al: Model for photon migration in turbid biological media, d Opt Soc Am (A) 4:423-432, 1987 2. Brazy JE: Effectsof crying on cerebral blood volume and cytochrome aa~. J Pefliao"112:457-461, 1988 3. Brazy JE, Lewis DV, Mitnick MH, et al: Noninvasive monitoring of cerebral oxygenation in preterm infants: preliminary observations. Pediatrics 75:217-225, 1985 4. Chance B, Leigh JS, Miyake H, et al: Comparison of time-resolved and -unresolved measurements of deoxyhemoglobin in brain. Proc Nati Acad Sci USA 85: 4971--4975, 1988 5. Cohen A, Wadsworth N: A light emitting diode skin reflectanceoximeter. Med Biol Eng 10:385-391, 1972 6. Cope M, Delpy DT, Reynolds EO, el al: Methods of quantitating cerebral near infrared spectroscopydata. Adv Exp Med Biol 222:183-189, 1988 7. Delpy DT, Cope M, van der Zee P, et al: Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 33:1433-1442, 1988 317
P. W. McCormick, etal. 8. Delpy DT, Cope MC, Cady EB, et al: Cerebral monitoring in newborn infants by magnetic resonance and near infrared spectroscopy. Scand J Clin Lab Invest (Suppl) 188: 9-17, 1987 9. Ferrari M, De Marchis C, Giannini I, el al: Cerebral blood volume and hemoglobin oxygen saturation monitoring in neonatal brain by near IR spectroscopy. Aflv Exp Med Biol 200:203-211, 1986 10. Giannini I, Ferrari M, CarpiA, etal: Rat brain monitoring by near-infrared spectroscopy: an assessment of possible clinical significance. Physiol Chem Phys 14:295-305, 1982 11. Hazeki O, Seiyama A, Tamura M: Near-infrared spectrophotometric monitoring of haemoglobin and eytochrome a, a3 in situ. Adv Exp Mefl Bid 215:283-289, 1987 12. Hazeki O, Tamura M: Quantitative analysis of hemoglobin oxygenation state of rat brain in situ by near-infrared spectrophotometry. J Appl Physio164:796-802, 1988 13. Hoffmann J, LObbers DW: Estimation of concentration ratios and the redox states of the cytochromes from noisy reflection spectra using multicomponent analysis methods. Adv Exp Med Binl 200:119-124, 1986 14. Jrbsis FF: Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 198:1264-1267, 1977 15. Mayevsky A, Nioka S, Chance B: Fiber optic surface fluoremetry/reflectometryand 3 I-p-NMR for monitoring the intracellular energy state in vivo. Adv Exp Med Biol 222:365-374, 1988 16, McCormick PW, Stewart M, Goetting M, el al: The use of nondiffusible optically active dye tracer for noninvasive quantification of cerebral blood flow. Stroke 20:134, 1989 (Abstract) 17. McCormick PW, Stewart M, Goetting MG, etal: Noninvasive cerebral optical spectroscopy for monitoring cerebral oxygen delivery and hemodynamics. Crit Care Med 19:89-97, 1991 18. McCormick PW, Stewart M, Goetting MG, etal: Regional cerebrovascular oxygen saturation measured by optical spectroscopy in humans. Stroke 22:596-602, 1991 19. McCormick PW, Stewart M, Ray P, el al: Measurement of regional cerebrovascular haemoglobin oxygen saturation in cats using optical spectroscopy. Neurol Res 13: 65-70, 1991 20. Nossal R, Bonner RF, Weiss GH: The influence of path length on remote optical sensing of properties of biological tissue. Appl Optics 28:2238-2244, 1989 21. Patterson MS, Chance B, Wilson BC: Time resolved
318
reflectance and transmittance for the noninvasive measurement of tissue optical properties. Appl Optics 28: 2331-2336, 1989 22. Seeds JW, Cefalo RC, Proctor HJ, etaI: The relationship of intracranial infrared light absorbance to fetal oxygenation. I: methodology. Am J Obstet Gyneeol 149: 679-684, 1984 23. Smith DS, Levy W, Maris M, etal: Reperfusion hyperoxia in brain after circulatory arrest in humans. Anesthesiology 73:12-19, 1990 24. Stern MD: In vivo evaluation of microcirculation by coherent light scattering. Nature 254:56--58, 1975 25. Takatani S, Cheung PW, Ernst EA: A noninvasive tissue reflectance oximeter. An instrument for measurement of tissue hemoglobin oxygen saturation in vivo. Ann Biochem Eng 8:1-15, 1980 26. Tamura T, Hazeki O, Nioka S, et al: In vivo study of tissue oxygen metabolism using optical and nuclear magnetic resonance spectroscopy. Annu Rev Physiol 51: 813-834, 1989 27. Tamura T, Hazeki O, Takada M, etal: Abserbance profile of red blood cell suspension in vitro and in situ. Adv Exp Med Biol 222:211-217, 1988 28. Taylor MB, Whitwam JG: The current status of pulse oximetry. Clinical value of continuous noninvasive oxygen saturation monitoring. Anesthesia 41:943-949, 1987 29. Wiernsperger N, Sylvia AL, J6bsis FF: Incomplete transient ischemia: a nondestructive evaluation of in vivo cerebral metabolism and hemodynamics in rat brain. Stroke 12:864-868, 1981 30. Wray S, Cope M, Delpy DT, et al: Characterization of the near infrared absorption spectra of cytochrome aa 3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochim Biophys Acta 933:184-192, 1988 31. Wyatt JS, Cope M, Delpy DT, et at: Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 2:1063-1066, 1986 32. Yamashita Y, Suzuki S, Miyaki S, etal: The neonate brain and breast imaging using NIR transillumination, in Chance B (ed): Photon Migration in Tissues. New York: Academic Press, 1990, pp 55-67 Manuscript received April 2, 1991. Accepted in final form July 16, 1991. Address reprint requests to." Patrick W. McCormick, M.D., 2213 Cherry Street, Suite 311, Toledo, Ohio 43608.
J. Neurosurg. / Volume 76 /February, 1992
