Recently, a review authored by Kish has examined a number of studies that sought to demonstrate reduced serotonergic function in ecstasy users (Kish 2002). The review is focused on comparisons of ecstasy users and non-users on measures of serotonin transporter site density using the radioligands McN-5652 and (-CIT. Kish argues that these studies are flawed both because of the unreliability of the ligands employed and because of methodological flaws in these imaging studies. In addition, he states that differences in ligand binding could also be due to gender differences and genetically-determined variance in serotonin transporter structure. The imaging study using McN-5652 performed by McCann and colleagues is explored in detail. While Kish addresses other methodological difficulties in this study, such as failure to match ecstasy users and controls on use of other drugs, Kish also discusses difficulties with the imaging method employed in this study. For instance, there is no published data on the test-retest reliability of McN-5652, and binding in areas with few serotonin neurons (such as cerebellum) may not accurately reflect binding to the serotonin transporter.

In describing the comparisons performed in this study, Kish notes that the data gathered from this study was so variable that McCann and colleagues log-transformed prior to conducting analyses. Hence estimates of the differences between ecstasy users and non0user controls based on charts and graphs in the paper are liable to underestimate the actual degree of difference between the two samples. If ecstasy users have lost up to 95% of their serotonin transporter sites, as suggested by Kish, it is surprising that there is little evidence of gross psychological or functional differences between ecstasy users and controls. While Kish accepts that the findings of McCann et al might demonstrate the existence of a decrement in serotonin function in polydrug users who use ecstasy, he does not believe that these findings can be used to estimate the severity of that decrement.

In concluding this review, Kish makes several recommendations for future comparisons of ecstasy users and non-users. These include the use of more specific ligands for measuring the serotonin transporter, greater efforts to locate and test a group of "pure ecstasy users" or at least to better match controls and ecstasy users on use of other substances, and reliance on forensic measures of past drug use rather than using self-report measures alone.

Direct quote from Kish 2002:

"...The percentage changes between mean control and ecstasy group values are not stated in the McCann publication, an omission that has caused some confusion in the literature. Since many readers do not appreciate that the mean values provided in the McCann figures have been log-transformed, the magnitude of the differences have been incorrectly calculated from the summary figure (Fig. 3, McCann et al., 1998a) as representing only a modest (e.g., 22%: Holland, 1999; 25%: Gamma, 2000) decrease in brain [11C](+) McN- 5652 binding in the ecstasy group. However, after taking into account the log-transformation of the Fig. 3 data, it is apparent that the average extent of loss of [11C](+) McN-5652 binding in the polydrug ecstasy users is actually quite severe, ranging from approximately 50% loss of binding in the cerebellum (a region with very low SERT levels) to a profound 85% depletion in the hypothalamus. The extent of reduction actually becomes even more striking if one accepts the argument of the investigators that ''. . . PET imaging with [11C] McN-5652 tends to underestimate the magnitude of reductions in 5-HT transporter density by about 50%.'' (McCann et al., 1999b). This would appear to suggest a global loss of about 75% to 95% of [11C](+) McN-5652 binding/SERT concentration in brain of ecstasy users who, presumably, do not display any clinical signs of neurotoxicity." (Kish 2002 p. 850)

Kish SJ (2002). How strong is the evidence that brain serotonin neurons are damaged in human users of ecstasy? Pharmacology, Biochemistry and Behavior, 71(4), 845-855 http://www.maps.org/publications//2002_kish_1.pdf

McCann U, Szabo Z, Scheffel U, Dannals RF, Ricaurte, GA (1998). Positron emission tomographic evidence of toxic effect of MDMA ("Ecstasy") on brain serotonin neurons in human beings. The Lancet, 352, 1433-1437. http://www.maps.org/publications//1998_mccann_1.pdf