Comment on Esler et al. () “Increased brain serotonin turnover in panic disorder patients in the absence of a panic attack: Reduction by a selective serotonin reuptake inhibitor”

The results presented in the study by Esler et al. (2007) demonstrate a marked increase in central serotonin turnover in people with panic disorder that does not appear to be related to differences in re-uptake transporter function (Hood et al. 2003, 2005), and a normalisation of 5-HT turnover by successful treatment with the selective serotonin reuptake inhibitor (SSRI) citalopram. As Esler et al. suggest, it is not possible to rule out the alternative explanation that the SSRI treatment effects are a consequence of repeated testing, but the decision to design the study in the way that they used was made, quite rightly, on ethical grounds. Nonetheless, these observations provide an important piece of the puzzle that is the serotonin hypothesis of panic.

There are two competing theories for the role of serotonin in panic (Hood et al. 2003). The first, the serotonin deficit theory, proposes that too little serotonin has an etiological role in panic disorder, and that SSRIs are effective as they increase synaptic serotonin. A process known as tryptophan depletion (TD) can be used to decrease 5-HT transiently by dietary restriction of the serotonin precursor tryptophan (Hood et al. 2005). TD increases the likelihood of panic following challenge in untreated and SSRI treated panic disorder, which is consistent with low synaptic 5-HT availability being associated with panic disorder. The competing theory, that there is too much 5-HT, suggests that SSRIs work by reducing 5-HT turnover via increasing inhibitory action at autoreceptors, thereby decreasing 5-HT and producing the therapeutic effect.

Esler et al. (2007) have speculated that higher 5-HT turnover is a consequence of increased raphe nucleus cell firing and equates to increased postsynaptic receptor occupancy by 5-HT. This interpretation of their results lends itself nicely to the serotonin excess theory. It is difficult, however, to reconcile the tryptophan depletion studies under these assumptions. The theory does, however, fit with data that there is a reduction, as measured with positron emission tomography (PET) scanning, in the binding of the 5-HT 1A receptor tracer [11C]-WAY 100635 in the raphe nucleus in patients with panic disorder; less receptors would mean less inhibitory control of the raphe (Nash et al. 2003).

A possibility is that the higher 5-HT turnover demonstrated in the Esler et al. (2007) study does not necessarily equate to increased receptor occupancy. If, however, patients with panic disorder are metabolising 5-HT at a greater rate than controls, then one might see high turnover but low effective occupancy; in keeping with the TD-supported hypothesis that a relative under-functioning through low levels of synaptic 5-HT are associated with panic. In this model, the SSRIs work to increase synaptic 5-HT and the reduction in turnover is due to a reduction in firing secondary to increased 5-HT in the synapse increasing inhibition through the 5-HT auto-receptors. Previous work using peripheral markers has suggested that antidepressants that block noradrenaline uptake increase noradrenaline synaptic efficiency in depression in this way (Hauger et al. 1988).

Despite substantial advances in our understanding of 5-HT in panic, we still do not know whether increased or decreased serotonergic throughput is associated with this illness. It is likely that further advances into this area will require similar methodologies to get closer to the aspect that it is the aim to measure. Assessing the specificity of high 5-HT turnover for panic disorder by measuring central 5-HT turnover in patients with anxiety disorders and depression may help to clarify the role of 5-HT in these prevalent and disabling conditions.