Serotonergic platelet markers of suicidal behavior--do they really exist?

Abnormal changes in platelets used as peripheral markers of central serotonergic functions are said to be associated with suicidality and depression, but this association has not been supported by consistent findings.

This review based on selected, often quoted publications in this area focuses firstly, on obviously contradictory findings and, secondly, on potential methodological flaws explaining these discrepancies.

The platelet 5-HT transporter has been found to have an inconsistent association with suicidality; furthermore, the specificity of imipramine for the 5-HT transporter is most likely low, since the number of platelet impramine binding sites has not been reliably associated with platelet serotonin uptake (Vmax). Significant changes of platelet serotonin content in suicidal individuals, as described in various studies, are most likely due to washout periods that are too short to eliminate the effects of a previous drug intake, or, in violent suicide attempters, due to blood loss and dilution. Similar methodological shortcomings might account for an often-reported elevated number of platelet 5-HT(2) receptor binding sites in suicidal individuals. In many studies, the results have not been sufficiently controlled for drug effects on platelet 5-HT(2) receptors, and associations of platelet 5-HT(2) binding with selective classifications of suicidal behavior are often generalized as further evidence for an association of platelet 5-HT(2) receptors with 'suicidality'. Finally, changes in platelet MAO-B-activity in suicidal patients have not been reproducibly found, and the impact of smoking on MAO-B activity has not been controlled in any studies.

Methodological flaws such as small sample sizes, insufficient matching criteria for controls, use of inadequate ligands in binding experiments, nonconsideration of comorbidity etc. and considerable methodological differences between studies limit their validity and comparability. It does not seem possible, at present, to integrate published findings and concepts into a plausible biological model of suicidality.