Photoacoustic Imaging of Stroke and Demyelination In Vivo

Lockwood, Joshua 1 ; Ringuette, Dene 1 ; Zhuo, Xun 3 ; Di Ciano-Oliveira, Caterina 3 ; Yucel, Yeni 3 ; Waspe, Adam 4 ; Drake, James 4 ; Levi, Ofer 1, 2

1. Institute of Biomaterials & Biomedical Engineering, University of Toronto; 2. The Edward S. Rogers Sr. Department of Electrical & Computer Engineering, University of Toronto; 3. Li Ka Shing Knowledge Institute of St. Michael's in Toronto; 4. The Hospital for Sick Children, Toronto

In Canada, the prevalence of multiple sclerosis (MS) increases 4.7 % per year, affecting 0.18% of the population in 2008.  Also, in Canada, there are over 50,000 new strokes per year which is an incidence rate of 0.14%.  Both diseases are significant burdens on the health care system with stroke costing economy more than $3.6 billion per annum.  Typically, clinical assessment in MS and stroke relies on magnetic resonance imaging (MRI).  However, MRI technology is both costly and requires long scan times.  Photo-acoustic imaging (PAI) is a promising imaging modality for animal and human clinical studies, combining optical excitation and ultrasound readout.  It exploits the highly distinct optical absorption spectra of the dominant tissue chromophores in the near-infrared window and the deep penetration of ultrasound, while decoupling the resolution losses due to combined forward and backward scattering.  Compared to MRI, PAI allows for the tracking of faster tissue dynamics at a lower cost, and with the potential to use genetically modify tissue constructs which can be tracked via changes in the tissue absorption spectra. This modality has advanced to the level that preclinical studies are now feasible with a tomographic reconstruction providing spatial and temporal resolution exceeding functional MRI for rodent imaging applications.  We demonstrate PAI as a powerful tool to evaluate brain diseases and therapies in rodent models of stroke and de-myelination, the latter being a defining feature of MS.  We induced stroke using a photothrombotic model and expanded the peri-infarct zone using an antithrombotic agent to mirror clinical intervention applied to stroke patients.  Our approach enabled stroke induction and the monitoring of both stroke progression and intervention without central nervous system invasion.  Stroke progression was tracked based on changes in the concentration of oxy-hemoglobin and deoxy-hemoglobin.  The photothrombotic agent used was inducible with green light which applied over a thinned portion of skull.  We investigated the profile of blood brain barrier disruption globally prior to stoke intervention using a small molecule near-infrared contrast agent.  We also used multi-spectral PAI to identify myelinated structures in rodent brains.  We contrasted mice genetically deficient in myelin basic protein, which become myelin deficient, with healthy mice in vivo.  Lipid rich structures were identified by exploiting a distinct lipid absorbance maximum.  We performed cardiac perfusions and whole brain extractions followed by ex vivo imaging to quantify the reduction in resolution attributed to the acoustic impedance of the skull and the vascular component of the signal.  Results were also validated with histological staining of myelin.  Our results support PAI as an investigative tool for rodent models of stroke and MS.