Feasibility study for brain perfusion analysis using changes of FiO2 (inspired oxygen percentage) and magnetic resonance imaging.

The term perfusion is usually employed to describe the process of nutritive delivery of arterial blood to tissues. In the case of the brain, perfusion is especially a crucial factor, since the intake of glucose and oxygen in this organ is continuous due to its extremely high activity and lack of reserves. Moreover, it is altered in many disorders, diseases and injuries, so it should be tightly controlled in order to ensure the proper functioning of the body. As previously mentioned, one of the most important parameters regarding perfusion is the level of oxygen that arrives to the tissue. The general term to describe oxygenation degree is the blood saturation, Sp02, which is defined as the ratio between the oxygenated carrier, oxyhemoglobin, and the total amount of hemoglobin present in the bloodstream.

Due to its importance in diagnosis, cerebral perfusion has been studied by means of several techniques, such as MRI and CT. However, in both cases perfusion is assessed using an external contrast agent, such as Gadolinium or iodinated compounds. By introducing external contrast agents, images do not rely upon intrinsic properties of the tissue, and injection process increases medical expenses, patient discomfort, and may involve certain risks.

Therefore, this project has been developed in order to analyze the feasibility of a novel approach to assess perfusion, using the own blood as an endogenous contrast agent, using magnetic resonance imaging (MRI). The approach relies on how abrupt changes in saturation affect to the mean intensity of certain zones in the MRI image. BOLD is a classical technique that follows a similar procedure, but it only studies differential perfusion changes during neuronal activation, instead of analyzing large changes over time. SpO2 alterations are achieved by modifying the fraction of oxygen inhaled, or FiO2. For that purpose, a gas mixer device supplies different mixtures of O2 and N2, with 0% and 100% oxygen content, respectively. Both mixtures are alternated in order to achieve an abrupt fall in saturation levels and a further recovery of those values, generating concatenated downwards and upwards slope in the saturation curve. Saturation values, which represent the input function to the system, are recorded using pulseoximeter devices. During those changes, image acquisition is performed in parallel, in order to record how mean intensity is affected in the image. Two different pulse sequences have been tested in the experiments, Spin Echo Planar Imaging (SE-EPI) and Gradient Echo (GE).

The results show a drop in the mean intensity values of the SE-EPI sequences, coinciding with the drop in the saturation curve registered by the pulseoximeters. No changes appeared with GE sequences.

Therefore, the project shows that under certain conditions, the own blood could act as an endogenous contrast agent for perfusion analysis using MRI, thus opening up a wide range of possibilities for the future.