Clinical Brain Monitoring with Time Domain NIRS: A Review and Future Perspectives

Near-infrared spectroscopy (NIRS) is an optical technique that can measure brain tissue oxygenation and haemodynamics in real-time and at the patient bedside allowing medical doctors to access important physiological information. However, despite this, the use of NIRS in a clinical environment is hindered due to limitations, such as poor reproducibility, lack of depth sensitivity and poor brain-specificity. Time domain NIRS (or TD-NIRS) can resolve these issues and offer detailed information of the optical properties of the tissue, allowing better physiological information to be retrieved. This is achieved at the cost of increased instrument complexity, operation complexity and price. In this review, we focus on brain monitoring clinical applications of TD-NIRS. A total of 52 publications were identified, spanning the fields of neonatal imaging, stroke assessment, traumatic brain injury (TBI) assessment, brain death assessment, psychiatry, peroperative care, neuronal disorders assessment and communication with patient with locked-in syndrome. In all the publications, the advantages of the TD-NIRS measurement to (1) extract absolute values of haemoglobin concentration and tissue oxygen saturation, (2) assess the reduced scattering coefficient, and (3) separate between extra-cerebral and cerebral tissues, are highlighted; and emphasize the utility of TD-NIRS in a clinical context. In the last sections of this review, we explore the recent developments of TD-NIRS, in terms of instrumentation and methodologies that might impact and broaden its use in the hospital.

A Systemic Review of Functional Near-Infrared Spectroscopy for Stroke: Current Application and Future Directions

Background: Survivors of stroke often experience significant disability and impaired quality of life. The recovery of motor or cognitive function requires long periods. Neuroimaging could measure changes in the brain and monitor recovery process in order to offer timely treatment and assess the effects of therapy. A non-invasive neuroimaging technique near-infrared spectroscopy (NIRS) with its ambulatory, portable, low-cost nature without fixation of subjects has attracted extensive attention. Methods: We conducted a comprehensive literature review in order to review the use of NIRS in stroke or post-stroke patients in July 2018. NCBI Pubmed database, EMBASE database, Cochrane Library and ScienceDirect database were searched. Results: Overall, we reviewed 66 papers. NIRS has a wide range of application, including in monitoring upper limb, lower limb recovery, motor learning, cortical function recovery, cerebral hemodynamic changes, cerebral oxygenation, as well as in therapeutic method, clinical researches, and evaluation of the risk for stroke. Conclusions: This study provides a preliminary evidence of the application of NIRS in stroke patients as a monitoring, therapeutic, and research tool. Further studies could give more emphasize on the combination of NIRS with other techniques and its utility in the prevention of stroke.

Cortical Spreading Depolarization (CSD) Recorded from Intact Skin, from Surface of Dura Mater or Cortex: Comparison with Intracortical Recordings in the Neocortex of Adult Rats

In cerebral cortex of anesthetized rats single waves of spreading depolarization (CSD) were elicited by needle prick. CSD-related changes of DC (direct current) potentials were either recorded from the intact skin or together with concomitant changes of potassium concentration with K⁺-selective microelectrodes simultaneously at the surface of the dura mater or of the cortex ([K⁺]_s) and in the extracellular space at a cortical depth of 1200 µm. At the intact skin CSD-related DC-shifts had amplitudes of less than 1 mV and had only in a minority of cases the typical CSD-like shape. In the majority these DC-shifts rose and recovered very slowly and were difficult to identify without further indicators. At dura surface CSD-related DC shifts were significantly smaller and rose and recovered slower than intracortically recorded CSD. Concomitant increases in [K⁺]_s were delayed and reached maximal values of about 5 mM from a baseline of 3 mM. They rose and recovered slower than simultaneously recorded intracortical changes in extracellular potassium concentration ([K⁺]_e) that were up to 65 mM. The results suggest that extracellular potassium during CSD is diffusing through the subarachnoid space and across the dura mater. In a few cases CSD was either absent at the dura or at a depth of 1200 µm. Even full blown CSDs in this cortical depth could remain without concomitant deflections at the dura. Our data confirmed in principle the possibility of non-invasive recordings of CSD-related DC-shifts. For a use in clinical routine sensitivity and specificity will have to be improved.

Characterization of motor and somatosensory function for stroke patients

In a pilot study, stroke patients with a lesion related to the motor system were studied using magnetoencephalography (MEG) and electromyography (EMG). The patients performed sustained finger movements for 30 s followed by 30 s of rest and 20 repetitions of this sequence in total. Task-related cortical signals derived from MEG were observed here at very different frequency scales. Slow signals below 0.1 Hz were extracted by independent component analysis and are associated with the sustained activation of the motor cortex, the dcMEG motor activation. MEG-EMG coupling phenomena in the 10-30 Hz range were analyzed using the imaginary part of coherency and are attributed to cortico-muscular coupling driving the muscles. Additionally a signal from the somatosensory cortex due to an electrical stimulation at the wrist, the N20m, was recorded as a physiological marker. Field maps and time series associated with the three types of signals are presented for one patient and one control subject as the signal quality of the patient data was not sufficient to achieve a group result. The feasibility of a comprehensive electrophysiological measuring and analysis procedure of the motor function for stroke research is demonstrated by the results.