Intraoperative Cerebral Autoregulation Assessment Using Ultrasound-Tagged Near-Infrared-Based Cerebral Blood Flow in Comparison to Transcranial Doppler Cerebral Flow Velocity: A Pilot Study John M. Murkin, MD, FRCPC,* Moshe Kamar, MD,† Zmira Silman, MSc,† Michal Balberg, PhD,† and Sandra J. Adams, RN* Objective: This was a pilot study comparing the ability of a new ultrasound-tagged near-infrared (UT-NIR) device to detect cerebral autoregulation (CA) in comparison to trans- cranial Doppler (TCD). Design: An unblinded, prospective, clinical feasibility study. Setting: Tertiary-care university hospital cardiac surgical operating rooms. Participants: Twenty adult patients undergoing cardiac surgery with cardiopulmonary bypass (CPB). Interventions: There were no clinical interventions based on study monitoring devices, but a continuous correlation analysis of digital data from transcranial Doppler (TCD) velocity was compared with a novel UT- NIR device and correlation analysis of change signals versus mean arterial pressure was performed in order to detect presence or absence of intact CA and for deter- mination of the lower limit of cerebral autoregulation during CPB. Measurements and Main Results: Similar and highly sig- nificant concordance (κ ¼ 1.00; p o 0.001) was demon- strated between the 2 methodologies for determination of CA, indicating good correlation between the 2 method- ologies. Intact CA was absent in 2 patients during CPB, and both devices were able to detect this. Conclusions: To the authors’ knowledge this is the first clinical report of a UT-NIR device that shows promise as a clinically useful modality for detection of CA and the lower limit of cerebral autoregulation. The utility of UT-NIR was demonstrated further during times at which extensive usage of electrocautery or functional absence of the transcranial window rendered TCD uninterpretable. & 2015 Elsevier Inc. All rights reserved. KEY WORDS: cerebral blood flow, CBF, transcranial Doppler, TCD, ultrasound-tagged near-infrared device, UT-NIR, cardiopulmonary bypass, CPB, cerebral autoregulation C ARDIAC SURGERY employing cardiopulmonary bypass (CPB) has been associated with a variety of adverse neurologic and systemic outcomes. 1 Spontaneous and procedural variations in either cardiac output or pump flow rate during CPB, as well as changes in PaCO 2 , mean arterial pressure (MAP), temperature, and cerebral metabolic rate, can produce alterations in regional and global cerebral blood flow (CBF) and may be etiologic in neurologic dysfunction. 2 A developing body of literature is demonstrating a strong correlation between perfusion below the lower limit of cerebral autoregulation (LLA) and associated major organ morbidity and mortality. 1,3,4 Although posing technical challenges, relative changes in CBF can be quantified reliably using transcranial Doppler (TCD) via insonation of the middle cerebral artery (MCA) to detect alterations in MCA flow velocity (MCAFV) during conditions under which MCA diameter remains constant. 5,6 A recently introduced and commercially available near-infrared (NIR)- based device incorporating ultrasound (US) phase shifting of emitted NIR photons (UTLight Flowmetry, Ornim Medical Ltd., Kefar Saba, Israel) also allows monitoring of relative changes in cerebral microcirculatory perfusion (cerebral flow index [CFI]). 7 As a “proof of concept” of the validity of CFI for con- tinuous noninvasive monitoring of cerebral perfusion, the authors undertook a study employing time domain-based correlation analysis between MAP and CFI and MAP and MCACFV as a means of assessing cerebral autoregulation (CA) and the LLA in patients undergoing cardiac surgical procedures employing CPB. Recognizing that TCD assesses flow velocity in major cerebral vessels whereas CFI monitors cerebral cortical micro- circulatory flow, the authors hypothesized that indices of microcirculatory perfusion, including presence of cerebral autoregulation and LLA during CPB, would be detected by UT-NIR flowmetry during conditions under which CBF was expected to undergo large changes (eg, changes in MAP, CPB pump flow rate changes, circulatory stasis, etc) and would correlate with changes in MCAFV. Because employment of a running correlation analysis between spontaneous changes in MAP and MCAFV has been demonstrated to detect presence or absence of CA in critical care and other clinical settings, 8 the authors further hypothesized that by using a similar analysis technique, UT-NIR flowmetry could be investigated to determine if MCAFV correlated with CFI the ability to evaluate CA and LLA during nonpulsatile CPB. 9,10 MATERIALS AND METHODS After receiving approval of the study protocol from the University Research Ethics Board (#17837; June 7, 2011) and obtaining written informed consent, 26 patients undergoing elective cardiac surgery with use of nonpulsatile CPB were enrolled in the study. Exclusion criteria included age o18 years, emergency surgery, stroke within preceding 3 months, and off-pump coronary revascularization. After induction of anesthesia using fentanyl, rocuronium, and sevoflurane, titrated to maintain the bispectral index (BIS) in a range between 40 and 60 throughout the intraoperative period, a From the *Department of Anesthesiology and Perioperative Medi- cine, Schulich School of Medicine, University of Western Ontario, London, Canada; and †Ornim Medical, Kfar Saba, Israel. Address correspondence to John M. Murkin, MD, Rm C3-112 University Hospital, 339 Windermere Rd, London, ON, N6A 5A5, Canada. E-mail: john.murkin@lhsc.on.ca © 2015 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2015.05.201 Journal of Cardiothoracic and Vascular Anesthesia, Vol 29, No 5 (October), 2015: pp 1187–1193 1187