Sir, we read with interest the paper on 'Phantom bite revisited'.1 The authors have not used a recognised diagnostic system and refer to 'mono-symptomatic hypochondriacal psychosis' which is no longer used in the psychiatric literature.

All the cases described were preoccupied with an imagined defect or had a minor physical anomaly and their concern was regarded as markedly excessive. They were clearly significantly distressed and handicapped. These are precisely the diagnostic criteria for body dysmorphic disorder 2 (BDD). If the beliefs are held with no insight, then an additional diagnosis can be made of a delusional disorder. In ICD-10, the term BDD or the older term 'dysmorphophobia' is not separately classified and is subsumed under hypochondriacal disorder.3

If however the beliefs are held with no insight, then an alternative diagnosis of 'Other persistent delusional disorder' is made. The preoccupation with a dental occlusion needs to be seen in the wider context of preoccupation in other patients with perceived defects such as slight flaws on the face, body features being too small or too big, hair thinning, wrinkles, spots, scars, asymmetry or lack of proportion. There is frequent comorbidity in BDD with depression, a significant risk of suicide 4,5 and dissatisfaction with cosmetic surgery. 6,7

The authors make a number of recommendations for treatment with pimozide (an anti-psychotic drug) which is not evidence based. There have been two randomised controlled trials of serotonin reuptake inhibitor with BDD.8,9

In short, the evidence favours the use of an SSRI in high doses for at least 12 weeks even in delusional patients. Furthermore Phillips10 has conducted a randomised controlled trial of pimozide augmentation of an SSRI in BDD or its delusional variant and found it to be no more effective than placebo. Phillips11 has also conducted a retrospective survey in BDD in which only 3% of 83 trials of an anti-psychotic (including pimozide) were of any benefit in BDD.

The authors also refer to building adaptive coping skills and overall functioning. We would agree, but this needs to be in the context of a testable theory and identifying the factors that maintain the preoccupation.

This can be done through a cognitive behavioural model12 for which there is an evidence base with two RCTs.13,14 The key issue in engagement of a patient with BDD or hypochondriasis is helping the patient to have a good understanding of what the problem is (rather what it is not) and validating the patient's experience. The patient is trying too hard to solve the wrong problem and their solutions have now become their problem.

Therapy therefore progresses in trying to identify the factors that maintain the preoccupation and distress. These might include selective attention and increased awareness of small occlusal discrepancies; ruminating about the sensation; the excessive monitoring of the sensation by checking and clenching of teeth and bruxism; treatment seeking and further pain leading to increased awareness in a vicious circle.

This in turn would lead to avoidance of normal activities. All of the maintaining factors have to be dismantled for the patient to become functional.

One of the authors of the paper, R G Jagger responds: We appreciate the interest in our article and in general agree with the views of Drs Veale and Chapman. However, although it is true that mono-symptomatic hypochondriacal psychosis is not a term currently used in the psychiatric literature, it was the term originally used by Marbach 15 . This original conceptualisation is still widely read and it was this historical term which we were 'revisiting'. There is indeed an overlap between body dysmorphic disorder (BDD) and 'phantom bite', with the shared feature of markedly excessive concern over a minor feature with consequent impairment in important areas of function.

However, an essential requirement for a diagnosis of BDD, as classified in DSM-IV2, is a preoccupation with appearance and it is important to note that none of the three reported patients had such a preoccupation.

As Drs Veale and Chapman point out, in ICD-10 both BDD and phantom bite fall into the category of 'hypochondriacal disorder' which subsumes a widely heterogeneous group of disorders. However, the point is that both DSM-IV and ICD-10 are over ten years old and more recently there has been a shift away from this categorical approach of classification.

Systems in which individuals are fitted into subcategories that are separate and mutually exclusive are being replaced by dimensional ones in which symptoms are grouped together within different symptom dimensions that can coexist to different degrees in individual patients 17, 18 . But, any further academic discussion here of psychiatric classification is probably not helpful to the busy dental practitioner who is trying to deal with a patient with phantom bite.

Whilst we agree that SSRI's have been used successfully in the treatment of related conditions such as BDD, we would not recommend that the general dentist take this approach but, as we emphasised in our paper, patients with phantom bite who do not respond to obvious conservative measures should be referred for psychiatric assessment and specialist management.

This may include CBT, but in practice it is often impossible to persuade patients even to consider this option. Also, as described in Case 2, they may be unwilling to take any medication, despite symptoms of severe major depression with suicidal ideation.

It is encouraging that there is an interest in these difficult to treat conditions. We hope that the major advances that have taken place over the last few years in the understanding of neuropsychiatric disorders will move us away from rigid classification towards a more rational basis for developing treatments19.