Non-invasive, Continuous Measurement of Cerebral Blood Flow During Progression to Brain Death
Rebecca Stafford1, Christian Arbelaez2, Ailis Muldoon3, Ona Wu4, Maria A Franceschini3, David A Boas5, David Greer2, Stefan Carp3, David Chung3, Brian Coffey1
1Boston Medical Center, 2Boston University School of Medicine, 3Massachusetts General Hospital, 4MGH Stroke Research Center, 5Boston University
Objective:

To study cerebral blood flow in a post-cardiac-arrest patient during progression to brain death using a non-invasive and continuous modality. 

Background:

Accurate and timely brain death testing is important to ensure high-quality neurologic and end-of-life care. While there are clear guidelines for brain death determination, the optimal timing for initiating testing remains uncertain. A continuous and non-invasive modality could help determine when cessation of intracranial blood flow has occurred and aid in determining the timing to initiate brain death testing. 

Design/Methods:

Optical data was obtained using a custom-built diffuse correlation spectroscopy (DCS) device, a continuous, non-invasive modality to measure cerebral blood flow. We measured cerebral blood flow in one patient using a custom-built, fiber-optical probe placed on the patient’s scalp that included collection fibers at two source-detector separations to detect scalp versus cerebral blood flow. We recorded continuous arterial line and ECG waveforms. Data were processed and analyzed using MATLAB to calculate a cerebral blood flow index. 

Results:

A 63-year-old woman presented after out-of-hospital asystolic cardiac arrest with extended down time requiring five rounds of CPR. Admission neurological exam showed GCS 3 with only bilaterally present pupillary reflexes. Initial CT head (post-arrest day 0) showed diffuse edema and near-complete effacement of the basal cisterns. We saw decrement of cerebral blood flow across post-arrest days 2-7, which was associated with total loss of brainstem reflexes and catastrophic hypoxic-ischemic brain injury with herniation on follow up CT imaging. Formal brain death testing was performed post-arrest day 6. Brain SPECT scan on day 7 showed no detectable blood flow to the brain, compatible with brain death. 

Conclusions:

DCS monitoring can capture continuous, relevant physiologic changes during progression to brain death in a single patient after cardiac arrest. These findings implicate a potential role for DCS as a noninvasive tool to determine timing for initiating brain death testing. 

10.1212/WNL.0000000000204757