One of the theories on the biology of panic disorder which received much attention is the locus coeruleus hypothesis. This hypothesis states that there exists an episodic overactivity of noradrenergic systems in panic disorder. Experiments with yohimbine form the cornerstone of this theory, and additional supporting evidence is provided by challenge studies with isoproterenol and treatment studies with clonidine. These studies are critically reviewed and it is argued that based upon the studies with yohimbine one cannot conclude that there is a specific noradrenergic abnormality in panic disorder, since yohimbine also affects dopaminergic and serotoninergic neurons in the brain.

In an alternative theory on the relationship between noradrenaline and anxiety it is postulated that increases in anxiety, irrespective of the diagnostic category, are related to increases in locus coeruleus activity.

A considerable amount of patients with panic disorder (25 – 30%) is refractory for the first line of treatment, mostly antidepressants, and for psychotherapy. Augmentation and combination strategies are discussed; and a protocol for the treatment of refractory patients with panic disorder is proposed with consists of five steps.

No patients is considered treatment refractory until they have had a trial with monoamine oxidase inhibitors.

In this review the results are discussed on the prediction of response to pharmacotherapy in panic disorder (PD) and social phobia (SP). From the literature it appears that in PD variables indicative of illness severity are predictors of non-response. In our own studies we found that non-responders to pharmacotherapy in PD are characterised by a higher score on the Blood-Injury subscale of the Fear Questionnaire, a higher plasma MHPG and a higher heart rate. In SP non-responders to pharmacotherapy were also characterised by a higher heart rate. They also appeared to have a higher blood pressure. Furthermore non-responders to pharmacotherapy in SP had higher scores on several psychometric scales, indicative of illness severity.

The purpose of this article is to give a comprehensive review of biological theories about the development of panic disorder.

Noradrenergic, serotonergic and GABA-ergic models of panic disorder are discussed, together with the role of peptides and neuroanatomical hypotheses.

The conclusion is that there is no unitary biological explanation of panic disorder.

An intergrative bio-psycho-social model seems for the moment the most usefull.