Mean transit time image--a new method of analyzing brain perfusion studies

Parametric imaging of cerebral vascular reserves. 1. Theory, validation and normal values

We describe the theoretical basis and practical details of a method for creating parametric images of the ratio of regional cerebral blood flow (rCBF) to volume (rCBV) from planar gamma-camera images following intravenous injection of any intravascular tracer and after correction for dispersion of the bolus before it arrives at the head. This ratio, which is the most sensitive index of the extent to which perfusion is being maintained by vasodilatation in patients with cerebro-vascular disease, indicates whether the limits of compensation have been, or are in danger of being, exceeded and identifies any ischaemic penumbra. Evidence is provided of the validity of the correction and the accuracy of calculation of the ratio. The range of normal values established is not significantly different from that previously reported using positron emission tomography. This technique has proved sufficiently sensitive to be clinically useful in patients with abnormalities of cerebral perfusion.

Parametric imaging in cerebral radionuclide angiography (RNA) by planar imaging improving presentation and objectivation of cerebral blood flow

Methods of parametric imaging of radionuclide angiography using parameters like appearance time, peak time, transit time, height of peak, arterial slope and area of inflow were developed and evaluated regarding their diagnostic meaning in 111 patients suffering from TIA or PRIND and in 30 normal persons. The meaning of these single parameters could shown depended on the specificity of the diagnostic question. Local cerebral blood flow can be estimated most favourably by parametric images of area of inflow whereas transit time is most promising as a diagnostic tool for evaluation of total cerebral blood flow classified with reference to severity of the perfusion disturbance. Appearance time is suited very well to estimation of collateral perfusion. Blood flow in great cerebral arteries could be seen well by non parametric imaging of radioactivity inflow in the brain supplying arterial vessels in the cranial floor. Applying a combination of the parametric images, the sensitivity for detection of disturbances of cerebral blood flow amounts to 0.91. A specificity of 0.88 and accuracy of 0.90 were found. The described combination of evaluation of RNA using various parameters is considered a well suited method for detection of disturbances in local and total cerebral blood flow by means of planar imaging.

Near infrared determined cerebral transit time and oxy- and deoxyhemoglobin relationships during hemorrhagic hypotension in the dog

The goal of the present study was to investigate the relationships among CBF, CMRO2, MTT and tissue/superior sagittal sinus hemoglobin saturations over a wide range of CPP using a canine model of hypotension. Microsphere-determined CBF was autoregulated in all tested cerebral regions over the 40-130 mmHg CPP range, but it decreased 50% around 30 mmHg. MTT, calculated by an indicator-dilution technique, progressively and proportionately increased over the 40-130 mmHg CPP range. The HbO2 content, measured in the same area by an optical method, parallely decreased. Around 30 mmHg CPP, MTT disproportionately lengthened (700% of control). CMRO2 maintenance was accompanied by a progressive decrease of tissue HbO2 and sagittal sinus hemoglobin saturation. In this framework, MTT measurement and tissue HbO2 monitoring by NIRS might give a sensitive evaluation of the hemodynamic reserve than CBF measurement in anesthesia and intensive care.

Evaluation of regional cerebral circulation based on absolute mean transit times in radionuclide cerebral angiography

In radionuclide cerebral angiography (RCA) the analysis of the activity-time curves recorded over the hemispheres following an intravenous bolus injection of a radioactive indicator permits the evaluation of absolute cerebral mean transit times (MTTcb). The inverse of MTTcb is an index of cerebral blood flow. The bolus parts of the cerebral activity curves were fitted with gamma-variate functions. The duration of the injection bolus and its subsequent dispersion through the cardiopulmonary circulation were corrected for using the activity curve over the aortic arch. In a control material of 52 subjects MTTcb was uncorrelated with age and equal to 3.2 +/- 0.67 s (means +/- 1 SEE) for the 37 subjects younger than 60. For the fifteen subjects older than 60, mean MTTcb showed a steep increase with age. The results were in close agreement with those obtained by other researchers using either correction of cerebral curves based on a physical flow model or direct injections into the internal carotid artery. Comparison of RCA with x-ray carotid angiography in 12 subjects was satisfactory. Repeated RCA examinations in nine subjects a few days apart yielded a reproducibility of 0.7 s for MTTcb. The RCA is non-invasive, rapid, easy to perform and without risk for the patient. The RCA is recommended for quantitative evaluation of major alterations in cerebral perfusion.

Model identification and estimation of organ-function parameters using radioactive tracers and the impulse-response function

Examination of the input-output events in functioning organs by the use of the impulse-response function (IRF) for a radioactive tracer is gaining more and more ground in nuclear medicine. This study summarizes the development of deconvolution analysis, laying special stress on the 'model-free' approach. System linearity and time invariance are discussed, and means of eliminating noise in IRFs originating from the input and organ-time-activity curves are outlined. Typical IRFs are illustrated by flow diagrams, time-domain curves, and their representation by Laplace transforms. The cases of nondiffusible and diffusible tracers as well as parenchymally extracted and transported substances are discussed. Methods for the derivation of models and for the calculation of physiologically important parameters from the IRFs are suggested.