Detection of Brain Hypoxia Based on Non-invasive Optical Monitoring of Cerebral Blood Flow with Diffuse Correlation Spectroscopy

Abstract

Diffuse Correlation Spectroscopy (DCS) non-invasively permits continuous, quantitative, bedside measurements of cerebral blood flow (CBF). To test whether optical monitoring (OM) can detect decrements in CBF producing cerebral hypoxia, we applied the OM technique continuously to probe brain-injured patients who also had invasive brain tissue oxygen (PbO₂) monitors.

Methods

Comatose patients (GCS < 8) were enrolled in an IRB-approved protocol after obtaining informed consent from the legally authorized representative. Patients underwent 6–8 hours of daily monitoring. Brain PbO₂ was measured with a Clark electrode. Absolute CBF was monitored with DCS, calibrated by perfusion measurements based on intravenous indocyanine green (ICG) bolus administration. Variation of optical CBF and MAP from baseline were measured during periods of brain hypoxia (defined as a drop in PbO₂ below 19mmHg for more than 6 min from baseline (PbO2 > 21 mmHg). In a secondary analysis, we compared optical CBF and MAP during randomly selected 12-minute periods of “normal” (> 21 mmHg) and “low” (< 19 mmHg) PbO₂. Receiver Operator Characteristic (ROC) and logistic regression analysis were employed to assess the utility of optical CBF, MAP, and the two-variable combination, for discrimination of brain hypoxia from normal brain oxygen tension.

Results

Seven patients were enrolled and monitored for a total of 17 days. Baseline-normalized MAP and CBF significantly decreased during brain hypoxia events (p<0.05). Through use of randomly selected, temporally sparse windows of low- and high-PbO₂, we observed that both MAP and optical CBF discriminated between periods of brain hypoxia and normal brain oxygen tension (ROC AUC 0.761, 0.762 respectively). Further, combining these variables using logistic regression analysis markedly improved the ability to distinguish low- and high-PbO₂ epochs (AUC 0.876).

Conclusions

The data suggest optical techniques may be able to provide continuous individualized CBF measurement to indicate occurrence of brain hypoxia and guide brain-directed therapy.

Introduction

In present-day “state of the art” clinical care, critically ill patients are often admitted to the hospital with limited brain injury only to be discharged with significant neurologic disability caused by brain damage acquired during the hospital stay. This unsettling scenario occurs despite painstaking and expensive monitoring in the intensive care unit; it arises, in part, because we lack straightforward bedside methods to continuously monitor cerebral blood flow (CBF), especially its adequacy during progression of post-insult secondary brain damage. If detected early, decrements in CBF can potentially be treated to avert brain infarction. Unfortunately, the lack of widely applicable techniques for cerebral perfusion monitoring has resulted in clinicians making therapeutic decisions directed to non-neurologic endpoints such as blood pressure¹ and PaCO₂² derived from population based data, with expectations that such interventions will have a desired effect on an individual patient’s brain perfusion.

Non-invasive optical monitoring of brain oxygenation holds potential to provide the necessary information to move from population-based to patient-based interventions. However, most examples of optical monitors currently available for clinical use (Near-Infra Red Spectroscopy, NIRS) are incapable of absolute measurements of perfusion and metabolism. Previous work³ comparing a commercial cerebral oximeter to invasive brain oxygen tension (PbO₂)⁴ monitoring demonstrated a correlation between relative tissue oxygenation measured with the cerebral oximeter (rSO₂) and PbO₂; however the confidence interval was wide, and the rSO₂ values were poor predictors of brain hypoxia measured with PbO₂. Previous reports by Rosenthal, et al.⁵ and Johnston et al.⁶ have also demonstrated that PbO₂ is primarily sensitive to cerebral blood flow linked to blood pressure⁶. Thus, a non-invasive tool to monitor cerebral blood flow may be superior to devices measuring tissue oxygenation for predicting brain hypoxia measured by PbO₂.

Diffuse Correlation Spectroscopy (DCS) and Diffuse Optical Spectroscopy (DOS) are promising noninvasive optical techniques⁷ which can provide continuous bedside readouts of relative changes in CBF and tissue oxygenation.^8,9 Here, in a group of neuroICU patients with invasive PbO₂⁴ monitoring, we first measured the transit of intravenously injected ICG boluses in brain tissue to calibrate relative changes in DCS signal to absolute CBF. We then tested the hypothesis that this novel noninvasive optical monitoring (OM) instrumentation can detect periods of brain hypoxia induced by insufficient levels of CBF. We focused on the ability of DCS to detect changes in brain tissue oxygen, given the aforementioned previous studies showing that cerebral blood flow is a major predictor of brain tissue oxygen tension.

Methods

The Institutional Review Board of the University of Pennsylvania approved all aspects of the study. All procedures performed were in accordance with the ethical standards of the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants’ legally authorized representatives. Comatose patients (Glasgow Coma Score ≤ 8) with traumatic brain injury (TBI), subarachnoid hemorrhage (SAH), intracerebral hemorrhage (ICH), or post-ischemic/anoxic encephalopathy (PIAE) who underwent invasive multi-modality neuromonitoring as part of their standard care in the Neuro-intensive care unit were eligible for enrollment in this observational study. Invasive neuromonitoring at our institution includes placement of a parenchymal intracranial pressure (ICP) monitor (Camino, Integra LifeSciences, Plainsboro, NJ, USA) and a brain tissue oxygen (PbO₂) probe (Licox, Integra LifeSciences, Plainsboro, MA, USA) into frontal lobe white matter through a quad lumen bolt (Hemedex, Cambridge, MA, USA). In some patients, a cerebral microdialysis (CMD) probe (M Dialysis, North Chelmsford, MA, USA) and a thermodilution CBF monitor (Bowman Perfusion Monitor, Hemedex, Cambridge, MA, USA) were also placed; however, given the variability in usage of CMD and CBF monitors across enrolled patients, we focused on the capability of non-invasive optical monitoring to detect episodes of decreased CBF associating with brain hypoxia in this initial validation study. Clinical physiological data (including blood pressure, heart rate, end-tidal CO₂, oxygen saturation, and data from intracranial monitors) were continuously recorded onto a bedside monitor (CNS Monitor, Moberg Research, Ambler, PA, USA) for subsequent offline analysis.

Optical CBF monitoring

CBF measurements were made using diffuse correlation spectroscopy (DCS).^7,8 The OM sensor was secured on the scalp over the frontal cortex area on the same hemisphere as the bolt (Figure 1). DCS estimates blood flow by quantifying rapid speckle intensity fluctuations of multiply scattered light through tissue induced by red blood cell motion^10–13 (Figure 2A). Specifically, DCS measures the normalized temporal intensity autocorrelation function, g₂(Δt) = ⟨I(t)I(t + Δt)⟩ / ⟨I(t)⟩², at multiple delay-times, Δt. Here I(t) is the detected light intensity at time t, and the angular brackets, ⟨⟩, represent timeaverages (in this study, an averaging time of 10 seconds was used). We employ a semi-infinite homogenous tissue model to derive a DCS blood flow index, F, from the decay of g₂(Δt) (Figure 2B). The DCS blood flow index, F, and its relative changes, have been validated to be proportional to tissue blood flow against a plethora of gold-standard techniques.^12,14 In our study, the relative changes in blood flow calculated by DCS were converted to “absolute” CBF values by calibration against a concurrent and co-located NIR measurement of the transit of intravenously injected indocyanine green (ICG) through brain tissue. Each monitoring day, a single time point estimate of absolute CBF was derived by monitoring cerebral and arterial indocyanine green (ICG) concentrations following intravenous bolus administration (0.1 mg/kg, injected in <3 seconds), as described previously by Diop et al.¹⁵ and He et al.¹⁶ and as illustrated in Figure 3. This dynamic contrast enhanced approach for measurement of CBF has been validated in preclinical studies¹⁵ and in healthy humans.¹⁷

Figure 1.

Configuration of optical monitoring probe and intracranial monitors. The optode was placed on the forehead adjacent to the intracranial monitoring bolt.

Figure 2.

Method of ICG-calibrated diffuse correlation spectroscopy (DCS) for absolute, real time CBF measurements. (A) An infrared light source is used to probe turbid media with moving particles (i.e., red blood cells; red disks at time t, light red disks at time t+Δt). Specifically, light propagates diffusively through the tissue along random walk pathways, and is scattered by moving red blood cells. The light scattered back from the tissue is measured at a detector placed adjacent to the source (source-detector separation is 2.5 cm). In our studies, the source/detector combination was incorporated into a single non-invasive optode patch. Light scattering by moving particles induces rapid (i.e., μs) temporal fluctuations in the detected speckle intensity, which are quantitatively characterized by the normalized intensity autocorrelation function. (B) Changes in the decay of the autocorrelation function over time are due to changes in CBF, which allows for a relative CBF index to be calculated over time. The rCBF index can be converted to absolute CBF by concurrent near infrared spectroscopic measurement of the transit of ICG through brain tissue using the same optode.15

Figure 3.

Overview of method used for ICG calibrated NIRS for absolute CBF measurement. (A) An intravenous bolus of ICG is given (0.1 mg/kg), and the temporal ICG concentration is measured in the arterial circulation via a custom pulse oximeter (or dye densitometer, with wavelengths of 804 and 938 nm that are optimal for ICG measurement) and in brain tissue via time-resolved DOS (using the optode with a source-detector separation of 3.2 cm, wavelength of 810 nm). The arterial and brain tissue ICG concentrations were sampled at 1 Hz. (B) Example brain tissue ICG and arterial impulse response function from a patient is shown. Tissue ICG concentration is equivalent to the convolution of the arterial impulse response (i.e., the arterial ICG concentration) and CBF×R(t), where R(t) is the fraction of ICG that remains in tissue at time t. Absolute CBF is calculated by deconvolving the tissue ICG concentration with the arterial ICG concentration to obtain CBF×R(t), and then taking the t=0 intercept (since R(0) = 1 by definition).

The custom-built OM instrument consists of time-resolved DOS (TR-DOS) and DCS modules.¹⁸ The TR-DOS light source is a commercial supercontinuum fiber laser (SuperK Extreme EXR-20, NKT Photonics Inc, Morganville, NJ, USA) that emits short white-light pulses (400–2400nm, seed pulse width ~5 ps) at a repetition rate of 78 MHz. The fiber laser is coupled through an acousto-optic tunable filter (SuperK Cross, NKT Photonics Inc) for programmable selection of specific wavelengths to deliver to tissue via a SuperK Connect fiber delivery system (FD7, NKT Photonics Inc). Hybrid single-photon sensitive photomultiplier tubes (PMA hybrid 50, Picoquant Photonics Inc, Berlin, Germany) connected to a time-correlated single-photon counting module (HydraHarp 400, Picoquant Photonics Inc) was employed for TR-DOS photon time-of-flight measurements (1 ps resolution). Note, the PMA hybrid detector was equipped with an electrically controlled shutter that was open only during TR-DOS acquisition.

The DCS light source is a continuous wave, long coherence length (≥8 m) 785 nm diode laser (IBEAM-SMART-785-S-WS with Smartdock fiber coupler, Toptica Photonics Inc, Victor, NY, USA) connected to a fiber-coupled electrically controlled shutter (OZ Optics, Ottawa, Ontario, Canada) for time-gated light delivery to tissue. DCS measurements of near-infrared light intensity autocorrelation functions (10 Hz sampling rate) were made with 4 arrays of 4 high-sensitivity single-photon counting avalanche photodiodes (Excelitas SPCM-AQ4C, Pacer LLC, Palm Beach Gardens, FL, USA) connected to a multiple-T 16-channel hardware correlator (Correlator.com, Bridgewater, NJ, USA) operating in burst mode.¹⁹ For more details about DCS and TR-DOS instrumentation, we refer readers to recent reviews.^8,20,21 Optical fibers couple the DCS and TR-DOS light sources and detectors to the OM sensor secured to the head. In this report, a DCS source-detector separation of 2.5 cm and a TR-DOS source-detector separation of 3.2 cm were used.

Statistical analyses

From enrollment to removal of the brain bolt, patients underwent 6–8 hours of daily optical OM. We assessed the ability of non-invasive optical CBF monitoring to detect CBF decrements associating with brain hypoxia using two analytic approaches. In the first approach, we identified discrete episodes of brain hypoxia defined as decrement to PbO₂ < 19mmHg^4,22,23 for >6 min from a baseline PbO₂ > 21mmHg for >6 min (Figure 4), and we compared these brain hypoxia episodes with OM changes in CBF and with changes in mean arterial pressure (MAP) (one minute time windows). A Wilcoxon signed-rank test was used to compare distributions.

Figure 4.

Method for identifying hypoxic events. The plot shows ~1 hour of continuous PbO₂ data from subject OM-14. An episode of brain hypoxia was defined as the average PbO₂ in a 6 minute period of sustained PbO₂ < 20 mmHg that followed a 3 minute period where PbO₂ transitioned from > 20 mmHg to < 19 mmHg. The pre-hypoxia baseline was defined as the average PbO₂ in a 6 minute period of time that occurred 6 minutes prior to the transition to brain hypoxia.

The second analytic approach evaluated randomly selected periods of ‘low’ (≤19mmHg) and ‘normal’ (≥21mmHg) PbO2 from all of the monitoring days. For each of the monitoring days, we randomly selected up to four 12 minute periods of ‘low’ PbO2 at least 120 min apart and up to four 12 minute periods of ‘normal’ PbO2, again separated by at least 120 min. Given this temporal separation, each period was considered independent of any other period: data collected in one window is unlikely to be correlated to that in another. Each period was thus considered independent of any other period.

Receiver Operator Characteristic Curves were plotted to assist in screening potential non-invasive variables which might predict low PbO2.

Results

Seven patients with GCS<8 were enrolled with the following diagnoses: traumatic brain injury (n = 3), intracerebral hemorrhage (n = 2), and post ischemic encephalopathy (n = 2). The entire cohort underwent concurrent noninvasive OM for a total of 17 days. Each measurement day was considered independent.

Overall, both MAP and DCS-CBF showed a significant correlation with PbO₂ (Figure 5) but with noticeable inter- and intra-subject variation. A total of 17 brain hypoxia events (as defined by the PbO₂ changes described above) were identified in 3 of the 7 patients studied (Table 1). During the brain hypoxia events, we observed (Figure 6) significant changes in MAP (p =0.0004) as well as CBF (p = 0.01, normalized to pre-event baseline). Further, since we performed simultaneous TD-DOS measurements, we were able to analyze whether brain hypoxia was associated with drops in regional oxygen saturation (rSO₂). Interestingly, rSO₂ values were not significantly associated with episodes of brain hypoxia on PbO₂ (AUC 0.46), with only a modest inverse correlation observed (Figure 5).

Figure 5:

Scatter plot of MAP (mmHg), CBF (ml/100g/min), and SO₂ (fraction) vs. PbO₂ (mmHg) for randomly selected windows 12 minutes long (as discussed in the text). Fit lines and 95% confidence intervals are shown in black. CBF is substantially more sensitive to changes in PbO₂ than either of the other variables.

Table 1:

Summary of patient demographics, diagnoses, and contributions of each patient to the various types of PbO₂ episodes

ID	Age	Gender	Diagnosis	Days Monitored	Brain Hypoxia Episodes	Low PbO2 episodes	Normal PbO2 episodes
3	29	Male	TBI	3	0	0	12
4	63	Female	ICH	2	2	4	2
5	38	Female	PIAE	1	0	0	4
6	33	Male	PIAE	4	10	16	15
12	49	Female	TBI	2	0	0	8
14	62	Male	ICH	4	5	16	13
15	27	Male	TBI	1	0	0	4

ICH intracerebral hemorrhage, PIAE post-ischemic/anoxic enceplopathy, TBI traumatic brain injury

Figure 6:

(A) Brain tissue oxygen measurements during hypoxic events (B) Mean arterial pressure (MAP) and (C) cerebral blood flow relative to baseline (rF =CBF/CBF_Baseline)] during hypoxic events. Median baseline CBF was 13.9 ml/100gm/min (IQR 8.6:16.5). MAP (p = 0.00035) and rF (p = 0.01) are significantly lower during hypoxic events. *** = p < 0.001, ** = p < 0.01.

Recall that our second analysis utilized randomly selected, temporally sparse windows of low and high PbO₂. Low PbO₂ episodes were identified in 3/7 patients (n = 36 episodes), while normal PbO₂ episodes were identified in all patients (n = 58 episodes) (Table 1). We found that both MAP and changes in CBF discriminated between periods of brain hypoxia and normal brain oxygen tension (AUC 0.761, 0.762 respectively) (Figure 7). We next employed logistic regression to derive a weighted linear combination of these parameters that optimized separation of the high and low windows (Log odds of low PbO₂ = −18.0202+0.1805 *MAP+0.2191*CBF, all coefficients significant at p<0.05). This combined MAP and CBF data-type improved our ability to discriminate between high and low PbO₂ episodes (AUC 0.876) compared to either variable alone.

Figure 7:

Receiver Operating Characteristic (ROC) curves for MAP and optically measured CBF to separate randomly and sparsely chosen windows of low and high PbO₂. MAP and absolute CBF were associated with low PbO₂ (ROC AUC 0.761, 0.762 respectively). Combining these variables using logistic regression analysis markedly improved the ability to distinguish low and high PbO₂ epochs (AUC 0.876). TPR-True positive rate, FPR-False positive rate.

Conclusion

Invasive PbO₂ monitoring is frequently used to guide care in patients with acute brain injury, and when combined with ICP monitoring, it may play an important role in providing personalized care which may affect outcome. Stiefel et al.⁴ showed an association between low PbO₂ episodes and outcome after traumatic brain injury, and this observation has been supported by other reports.^23,24 Unfortunately, due to multiple factors, invasive neuromonitoring is only feasible for a small subset of patients with severe acute brain injury. Our data support the notion that noninvasive optical measurements of CBF may provide a useful surrogate indicator of brain hypoxia, thus enabling the identification of physiological states that are associated with secondary brain injury in patients for which it is not possible to place invasive monitors. While encouraging however, given the small sample size and lack of an independent test set, we urge the reader to be cautious about the generalizability of these findings. More work with larger samples sizes is needed.

To summarize, we found that reductions in both optical CBF and blood pressure predicted episodes of brain hypoxia. Changes in brain tissue oxygen (PbO₂) reflect the combined effect of changes in both cerebral blood flow, blood pressure, and levels of dissolved free plasma oxygen^5,6. For example, hyperventilation produces decrements in PbO₂ by reducing CBF, while hyperoxia⁵, transfusion²⁵ and increasing blood pressure^5,6 can all cause an increase in PbO₂ through multiple complementary mechanisms. Overall, these observations are consistent with our findings, and support the view that reductions in PbO₂ reflect multiple potentially adverse conditions, including reductions in cerebral perfusion and factors (such as reduced blood pressure) that contribute to low CBF states. Interestingly, we found that regional oxygen saturation (measured non-invasively by concurrent TD-DOS) did not predict brain hypoxia (i.e., low PbO₂). Although counter-intuitive, this finding partially agrees with a previous study comparing invasive PbtO₂ to measurements of rSO₂ with a commercial NIRS instrument.³ While rSO₂ and PbO₂ were correlated in that study, the rSO₂ changes were not sufficiently sensitive to reliably detect episodes of brain hypoxia. The decreased sensitivity of NIRS compared to DCS may be due to the fact that NIRS and invasive brain tissue oxygen monitors are measuring oxygenation in different brain compartments (blood and interstitial fluid, respectively), or that NIRS rSO₂ requires an assumption on the arteriovenous admixture: the signal is dominated by capillaries and venules.⁹ Assuming no change in tissue blood volume, the total blood flow in each vascular compartment is equal and therefore the precise compartmental contributions to the DCS signal will not influence the measurement. When blood volume is changing, high-temporal resolution DCS has demonstrated that pulsatile flow dominates the signal, suggesting that the arteriole contribution dominates DCS blood flow measurements.²⁶ Note that NIRS and DCS can provide complementary information on hypoxic brain states, with NIRS more faithfully reporting episodes due to systemic hypoxia or increased cortical cerebral metabolic demand and DCS reporting episodes caused by decreased cerebral perfusion. Further work is required to clarify these issues.

The observation that both MAP and noninvasive CBF predict brain hypoxia episodes suggests that our patients had significant impairments of cerebral autoregulation. Further work is needed to ascertain whether similar results will be seen in patients with less severe forms of brain injury, e.g., wherein autoregulation is relatively preserved. Our results also point to the possible use of continuous DCS as a means to extract a real time, direct, flow-based index of pressure autoregulation, i.e., similar to autoregulation indices that are indirect and employ ICP²⁷, PbO₂²⁸, and NIRS²⁹. In a recently published study in healthy volunteers, we demonstrated the potential utility of DCS to monitor cerebral autoregulation,²⁶ and we plan to investigate the feasibility of DCS to continuously measure autoregulation in brain injured patients.

The noninvasive OM we used is a promising technology for neurocritical care. DCS is an alternative method for assessing CBF that has light penetration properties similar to near infrared spectroscopy (NIRS) cerebral oximeters (e.g., Casmed FORE-SIGHT Elite, Covidien INVOS 5100C, Masimo O3, Hamamatsu NIRO, etc.). DCS extracts a blood flow index from measurement of temporal fluctuations in the reflected light intensity primarily caused by moving red blood cells (Figure 1). However, while the light utilized for NIRS, time-domain optical bolus tracking, and DCS is similar in wavelength and tissue penetration, there are differences in in depth sensitivity. Continuous illumination NIRS is most sensitive to superficial tissues (e.g., skin, scalp), while the additional information provided by both time-domain and dye bolus measurements permit greater selective sensitivity to deeper tissues. DCS utilizes a fundamentally different contrast... moving red blood cells instead of hemoglobin absorption- and its depth sensitivity is correspondingly weighted by the blood flow at each depth. In the brain, blood flow is many times higher than that of the skull/scalp, increasing DCS sensitivity to cerebral blood flow.³⁰

Our data suggest that DCS can likely detect an anaerobic condition, and, notably, it does not rely on assumptions regarding microcirculatory arteriovenous admixture⁹ as is the case with traditional NIRS cerebral oximeters. NIRS oximeters reflect an uncertain mixture of artery, capillary, and venous blood, and the precise arterial and venous contributions can vary within and across patients.³¹ These assumptions limit establishment of quantitative NIRS oximetry saturation thresholds for ischemia. Relative and absolute quantitative CBF measurements with DCS, however, are not subject to this limitation, because the technique is based on assessment of erythrocyte movement rather than hemoglobin-based light absorption, and total blood flow in each microvasculature compartment must be the same, in the absence of change in tissue blood volume. Future work will be required to confirm our observations and establish ischemic CBF thresholds, e.g., by expanding study populations and utilizing more sophisticated statistical models to account for potential intra-subject correlations.

The initial generation of DCS technology for assessing relative CBF has been validated in multiple contexts. Initial proof of concept evaluation entailed extensive studies in tissue phantoms³² wherein the medium’s viscosity and therefore thermal motion of scatterers could be controlled. In subsequent reports, investigators compared DCS measurements of flow variation to other standards including Doppler ultrasound,³³ laser Doppler,³⁴ and perfusion MRI.³⁵ In rodents, DCS detected hyperemia due to hypercapnia^36,37 and ischemia due to cardiac arrest³⁷ with appropriate changes in OEF and CMRO₂.³⁷ DCS-detected changes in CBF have been reported in still other animal studies,^12,38 including comparison to contrast enhanced time-domain CBF measurements¹⁵, and also with stable xenon enhanced CT in SAH patients¹⁴ and MR-based validation in children.³⁹ Overall, these and numerous other uncited validation studies demonstrated that DCS measurements of blood flow variations in humans are in reasonable agreement with theoretical expectation and with other measurement techniques. Our results presented here (e.g., Fig. 6), corroborate these findings and further support the study of DCS-detected changes in CBF to inform clinical care.

Incorporation of a time-domain near-infrared analysis with ICG intravenous injection has enabled quantitative calibration of absolute CBF for DCS. This calibration approach has been validated in a preclinical study in neonatal piglets versus perfusion CT¹⁵ and in healthy humans with MR arterial spin labelling CBF assessment,¹⁷ though more work remains. Our data using a semi-infinite medium calculation method adds to this literature by assessing absolute CBF in comparison with invasive PbO₂. Future work will include the use of layered models to remove possible signal contamination from the scalp;⁴⁰ we, and others, are exploring adaptations of these techniques suitable for critically ill patients with invasive monitors. Moreover, the stability of the DCS calibration for continuous absolute CBF monitoring with DCS has been tested via a second absolute CBF measurement with ICG injection roughly four hours after the calibration measurement.¹⁶ This test confirms stability of calibration over at least 4 hours of monitoring. This development of quantitative CBF monitoring, combined with assessment of SaO₂ and tissue O₂ saturation, additionally provides for the possibility of a monitor of quantitative CMRO₂ and OEF. Quantitative CBF monitoring by DCS with ICG-calibration is much less well studied than relative measurements of CBF with DCS; as further results emerge, it may prove to offer additional benefit to clinical decision making.

Our data suggests that optical techniques can provide continuous individualized regional CBF measurement to guide brain-directed therapy to avoid anaerobic conditions. Given our small sample size, the need for expanded clinical studies of non-invasive tools is apparent, both to improve the care of those who currently undergo invasive monitoring and also care for other patients, without invasive monitoring, but at risk of neurologic deterioration.