This project is concerned with characterizing and improving the delivery of pharmacologic and diagnostic agents to the central nervous system. (1) A small iodinated compound, iopamidol (MW 777Da) was evaluated for its ability to track by CT in real time the in vivo distribution of both small and large MW compounds delivered to cerebral white matter by convection-enhanced delivery (CED). 97 microliters of infusate containing iopamidol and 14C-labeled sucrose and 70 kDa dextran were infused at a flow rate (qv) of 0.5 to 1.0microliter/min into primates, and volumes of distribution (Vd) were measured both by quantitative autoradiography (QAR) and CT immediately at the end of infusion. Differences between QAR and CT-derived Vds ranged from 7.5 to 19.7% and were not significant. Diffusion-convection computations revealed that iopamidol-based CT Vds provided estimates for 70 kDa dextran Vds with an expected absolute error <20% for infusion volumes (Vi) falling below the line Vi=600 qv. Iopamidol Vd estimates serve only as a measure of the outer bound of distribution for reactive or highly permeable molecules. (2) Viral distribution was investigated by MRI and fluorescence techniques. Combidex particles were used to estimate by MRI the Vd for a 20-micron adeno-associated virus acting as a GFP vector and administered by CED. The associated transfection volume was measured by quantitative imaging fluorescence (Fuji FLA5000). Transfection volume apparently exceeded CED spread suggesting post-viral uptake spread or thresholding artifact. (3) Analysis of the distribution of muscimol following 1.4 microliter/min high-flow-rate CED into primate hippocampus was undertaken. 18mm retrograde distribution along the infusion catheter was shown consistent with expected backflow lengths computed from a poroelastic model of brain tissue. Radial distributions in transverse planes at maximum distance from the cannula tip were shown consistent with cylindrical diffusional transport; regression of theoretic functions to data yielded an estimate of the local clearance rate of .001 min-1 (Rsq=.989). Forward spread from the cannula tip was quantitatively consistent with only a small fraction of the infusate crossing the hemispherical tissue interface at this location. For an infusate concentration of 0.125mM, the maximum range of muscimol activity either along the ray extending forward of the catheter tip or along transverse plane radii are less than 10 mm. (4) Finite element simulation has been introduced to guide delivery protocols of ibotenate plus muscimol to hippocampus.
