The purpose of this study was to evaluate the effects of age on D2 receptor binding with 123I-iodobenzofuran (IBF) SPECT. METHODS Subjects were 40 healthy volunteers (age 19-83 yr), including 6 who had test/retest studies. Scans were acquired with a triple-head SPECT camera 3 hr postinjection of IBF (300 MBq). Striatal regions (caudate and putamen) were defined by two different region-of-interest (ROI) sets consisting of large volumes [(CLVs), 2.2 and 6.6 m] and small volumes [(SVs), 0.6 and 1.3 ml]. D2 binding (Rv=V3/V2) was quantified using our previously proposed multilinear regression technique. Effects of age on D2 binding were evaluated by fitting linear, exponential and logarithmic models. RESULTS The mean Rvs were 26% lower than LV for both putamen and caudate than the corresponding values from the SV due to the partial-volume effect. Although the identifiability of Rv using SV deteriorated slightly, the test/retest reproducibility of Rv measurements was equally excellent for LV and SV. The mean Rvs were 11% higher for putamen compared with those for caudate. D2 binding declined significantly with age (p < 10(-5)) for all three models. The nonlinear models were slightly superior to the linear model in describing the relationship between Rv and age. In these models, D2 binding declined with age, equally for caudate and putamen at 7%-13% per decade; the decline was progressively smaller with age. CONCLUSIONS IBF SPECT permitted reliable measurements of D2 binding in the caudate or putamen separately using small ROI volumes that significantly improved the quantitation loss from the partial-volume effect. Our results agreed with previous PET and postmortem findings of D2 binding losses with age. However, these age effects may be nonlinear. Age-related changes in D2 binding must be taken into consideration in clinical IBF SPECT investigations.