Summary:The effects of varying the data collection time on the calculation of cerebral blood flow and distribution volume via the integrated projection technique were studied in four human subjects. The significance of these results in terms of the limitations of the single compart ment model for 150-water was explored using computer simulations. The simulations helped to account for causes for the variations seen in blood flow and distribuOver the last few years, several general methods have emerged to quantitate local cerebral blood flow (lCBF) with the use of positron emission to mography (PET) (Frackowiak et al. , 1980; Huang et al. , 1982;Raichle et al. , 1983). Cerebral tissues de pend on ICBF for delivery of nutrients and for the removal of metabolic products, and therefore, quantitative measurement of ICBF provides an im portant way for assessing local brain function. Most methods currently being used with PET employ 150-water as the tracer and are based on a single compartment model for the calculation of ICBF. Recently, investigators have noticed that the ICBF estimate is a function of the total data collection time (Raichle et al., 1983). These observations raise questions about the adequacy of the single com partment model for describing the transport ki netics of 150-water in brain tissue (Huang et al., 1983). The present study addresses this important issue.
13tion volume as a function of the data collection time. Two different compartmental models were explored for better quantitation of blood flow and distribution volume.