Abi-Dhargham et al. (Eur. Psychiatry 20 (2005) 15) comprehensively review the contribution of brain imaging to understanding the mechanism of second generation antipsychotic drugs. The review touches on our findings showing that all the second generation drugs we studied with 123I epidepride single photon emission tomography (SPET) preferentially block D2/D3 dopamine receptors in the temporal cortex, vs. striatum (Pilowsky et al., 1997; Bigliani et al., 2000; [Reference Stephenson, Bigliani, Jones, Mulligan, Acton and Visvikis5]; [Reference Bressan, Erlandsson, Jones, Mulligan, Flanagan and Ell2]). They go on to state these findings must be viewed ‘with caution’ since the ‘simple ratio method’ was used to determine specific D2/D3 binding in striatum and temporal cortex. It is important to correct factual errors that by unchallenged repetition in respected journals, gain credibility. We addressed the issue of inaccuracies associated with the ratio method by simulation and experiment 2003 [Reference Erlandsson, Bressan, Mulligan, Ell, Cunningham and Pilowsky3], and showed that though the ratio method could underestimate striatal D2/D3 receptor occupancy, it was accurate in the temporal cortex for the timescale of an epidepride SPET experiment. The short timescale of 11C FLB 457 PET experiments conversely could lead to an underestimation of temporal cortical D2/D3 receptor occupancy in this area, and a mistaken acceptance that there is no difference in striatal vs. limbic cortical occupancy by second generation antipsychotic drugs, especially when two different PET ligands are used to estimate occupancy in each area. PET studies addressing this issue use a long-lived isotope (18F or Br76), and label D2/D3 receptors in striatal and extra-striatal regions at the same time in the same experiment [4,7]. These studies report similar results to our own. We also used reference tissue modelling and 123I epidepride SPET to show the chemically distinct atypical antipsychotic drugs amisulpride and risperidone preferentially block D2/D3 receptors in limbic cortical, over striatal regions [1,2]. Furthermore, within the striatum, D2/D3 receptor occupancy by risperidone and amisulpride was higher in the head of caudate (limbic cortical associative inputs) than putamen (mainly motor connections) [Reference Stone, Bressan, Erlandsson, Ell and Pilowsky6]. It is worth noting that these findings are dose-dependent for drugs with relatively high affinity for the D2 receptor, so that ‘limbic selectivity’ disappears at higher plasma levels, and all PET and SPET D2/D3 occupancy studies should be viewed with this in mind.

Notes added in proof.