Covert stroke occurs in one out of fourteen patients during or shortly after surgery, and may result in long-term disability. Fortunately, stroke that occurs during non-cardiac surgery is most commonly caused by inadequate blood flow to the brain and is, therefore, preventable if it can be detected early. Current clinical tools used to monitor the brain during surgery do not have the accuracy nor the spatial coverage - they only monitor one small region of the brain. In this study, the investigators plan to apply a cutting-edge optical device, tr-fNIRS, to monitor the whole brain during shoulder surgery. The primary aim is to determine any regional differences in cerebral oxygenation (ScO2) and cerebral autoregulation (CA)between brain regions during surgery and especially during various physiological challenges, such as hypotension. The investigators hypothesize that certain brain regions are more likely to develop cerebral desaturation and impaired CA, and are more prone to brain injury than the frontal lobe region which is the traditional monitoring site. The investigators also hypothesize that cerebral desaturation (or hypoxic injury) events correlate with adverse postoperative neurological outcomes such as covert stroke, overt stroke and/or postoperative delirium.

Perioperative stroke is a significant complication after surgery. Unfortunately, covert stroke occurs in one out of fourteen elderly patients after non-cardiac surgery and is associated with an increased risk of long-term cognitive decline. Further, perioperative overt stroke is associated with an 8-fold increase in perioperative mortality, prolonged length of hospital stay, and decreased quality of life. For shoulder surgery performed in the beach chair position, due to systemic hypotension and reduction in cerebral perfusion pressure, cerebral desaturation events and impaired cerebral autoregulation can occur in up to 80% of patients. Most perioperative stroke in non-cardiac surgery is ischemic and is related to brain hypoperfusion. Importantly, brain hypoperfusion is potentially modifiable with simple measures such as increasing systemic blood pressure (BP) if brain ischemia is identified early enough. However, deliberate hypertension is not devoid of risk and there is currently no effective monitor that can detect brain hypoperfusion during surgery.

Cerebral oxygenation has been used clinically as a measure of adequate brain perfusion during surgery. An early clinical trial at Western found that monitoring cerebral oxygenation was associated with fewer cases of major organ dysfunction in cardiac surgery patients. Despite this promising initial finding, current commercial cerebral oximeters have a number of limitations that prevent reliable perioperative neuromonitoring, including non-specificity due to signal contamination from extracerebral tissue, especially during administration of vasoconstrictors or hypothermia. In addition, current commercial cerebral oximeters only have two channels to monitor the frontal lobe regions (i.e., anterior cerebral artery territory). This monitoring strategy assumes that cerebral oxygenation is homogenous across different brain regions so that measurements from the frontal lobe regions can be used clinically to represent the adequacy of global brain oxygenation. Such a limited spatial coverage may result in undetected stroke, despite the patients having apparently normal cerebral oxygenation in the frontal regions throughout surgery. The current research project employs multi-channel tr-fNIRS to address these limitations, with the goal of timely detection and prevention of ischemic brain injury.

Multi-channel tr-fNIRS (time-resolved (tr) functional near infrared spectroscopy (fNIRS) (tr-fNIRS)) is an emerging brain-imaging technology that was originally developed to explore changes in cerebral oxygenation generated by cortical neuronal activity during various cognitive tasks such as speech, sensory and motor functions, and emotion. This is because tr-fNIRS can measure such subtle cerebral oxygenation changes in milliseconds (up to 100 ms) that accompany neuro-activation such as finger-tapping. The investigators has built several tr-fNIRS devices and have the expertise of adapting the full head coverage tr-fNIRS device to perioperative neuromonitoring. Furthermore, tr-fNIRS operate by sending short (picosecond) pulses of light into the head and precisely measures the time of travel of each photon in the tissue. Since there is an equivalence between time and distance, photons that are detected right after the pulse have only probed the extracerebral layers (scalp and skull) while photons that are detected long after (1-2 ms) the pulse have travelled deep into the brain tissues. The investigators have recently shown that this approach reduces the signal contaminations of the extracerebral layers from 80% with current commercial NIRS devices to less than 8% with tr-fNIRS. In this study, the investigators will employ a state-of-the-art full head coverage tr-fNIRS device to monitor the entire brain, as opposed to only select regions (such as the limited capabilities of the current cerebral oximeters) in the perioperative setting. Together with in-house analysis algorithms, the full head coverage tr-fNIRS can detect specific brain regions at-risk of ischemic injury with a high degree of certainty because of greater spatial resolution (in cm) and less signal contamination from extracerebral tissue.

All study participants will be recruited and consented adhering to the local ethics guidelines. For all study participants, the surgical and anesthetic management of the patients will be conducted in a standard fashion and will not be altered in this study. The exception is that tr-fNIRS will be used to monitor regional brain oxygenation from anesthesia induction to completion of surgery. The surgeons, anesthesiologists, and nurses will be blinded to the monitor/measurements during the procedure. No intervention will be administered based on the results of the tr-fNIRS.

Inclusion Criteria:

• Adult patients (age ≥ 18 years old)
• Scheduled to have elective shoulder surgery in the beach-chair position under general anesthesia
• Provide informed consent

Exclusion Criteria:

i) have skin/scalp lesions that preclude the application of fNIRS device to the head ii) lack of written consent iii) emergency surgery

Shoulder surgery patients were chosen as the study population because of the high frequency of intraoperative hypotensive and cerebral desaturation events that may occur during this procedure, and the unique physiological changes such as impaired cerebral autoregulation that accompany the beach chair surgical positioning. This unique but commonly encountered operative condition create ischemic challenges and injury to the brain and provide opportunities to assess the spatial and temporal differences of cerebral desaturation patterns experienced during hypotensive and other physiological challenges, and to refine the tr-fNIRS device for perioperative applications.