A compartment model for fluorometric indication of tissue perfusion

A theoretical analysis incorporating a multicompartmental model for dye distribution is applied to describe the relationship among perfusion, dye kinetics, and fluorescence readings. Errors in calculating perfusion from fluorescence are related to dye concentrations and to measurement errors. These error sources are minimized by selecting the measurement time. The alternatives of measurement during wash-in or wash-out of dye are compared, as well as the alternatives of introducing dye by bolus injection or by constant infusion. Compensation of wash-in measurements for differences in skin pigmentation is accomplished by the matching of skin optical properties between incident and fluorescing wavelengths. A laboratory study at 80 measurement sites (from ten graded-perfusion canine flaps) demonstrated a correlation of wash-in and wash-out perfusion data ranging from 0.88 to 0.96 at typical levels of fluorescein in blood. Since wash-in can be completed in a matter of a few minutes, this is likely to be preferable in the clinical setting to wash-out, which can take much longer to complete.