Editorials

Medically unexplained neurological symptoms

The risk of missing organic disease is low

See p 582

A psychological component exists in all illness, contributing a variable amount to the clinical presentation. At the benign end of this range are patients who describe their symptoms in more florid terms than seem to be justified. At the malignant end is malingering and the Munchausen syndrome. Between these extremes are a range of patients who present with non-organic signs or symptoms, with or without underlying neurological disease, usually called hysterical elaboration or hysterical conversion disorder.

Again a range of clinical presentation exists. If patients present with apparent non-organic signs or symptoms and are later found to have an underlying disease which might account for some or most of their original problems this is perceived as a hysterical elaboration of the underlying deficit. For example, a patient who seems to have a hysterical paraplegia might be found to have white matter changes on magnetic resonance imaging and oligoclonal bands in the cerebrospinal fluid. Patients already known to have a neurological condition might develop non-organic signs or symptoms, sometimes related to the original condition, but often distinctfor example, the development of pseudoseizures in a patient with epilepsy from childhood associated with an abnormal electroencephalogram. Occasionally patients present with clearly hysterical signs or symptoms but later develop a recognisable neurological disease, which in retrospect might have explained some or all of the original presenting features. Finally, there are patients with hysterical signs and symptoms with no underlying organic disease and in whom no organic disease develops over many years of follow up.

Crimlisk et al have looked at a small part of this rangepatients who presented with purely motor symptoms for which no cause was found on clinical evaluation and investigation (p 582).¹ Only three out of 64 subjects followed for a mean of six years were subsequently diagnosed as having a neurological disorder that could wholly or partly explain the presenting symptoms. One of these patients had a paroxysmal hemidystonia, always a difficult diagnosis and a condition not fully characterised at the time of the original presentation. In one patient the eventual diagnosis was dystrophia myotonica and in another a spinocerebellar degeneration. These conditions are usually diagnosable at presentation, but communication difficulties apparently confused the initial assessment. However, this serves only to emphasise the authors' conclusion that patients who present with medically unexplained motor symptoms and who have been properly evaluated clinically and investigated are unlikely on follow up to show evidence of an underlying disease that might have accounted for the presenting features.

Crimlisk et al conclude that repeated investigations for the same problem are unprofitable. In common with other series they found that a good outcome is associated with a short history and warn that conversion disorders do not protect patients from developing other diseases. A previous occurrence of a conversion disorder may delay the diagnosis of another condition. Although these findings can probably be applied to the whole range of non-organic disease, we do not know whether patients with pseudoepileptic seizures, hysterical pain syndromes, hysterical blindness, or non-organic sensory impairment would follow the same pattern.

At least two other series on conversion disorders have been reported from the National Hospital for Neurology and Neurosurgery,^2-4 the most influential being that of Slater in 1962. He analysed 112 patients seen in 1951, 1953, and 1955 ^{2 3} and followed 85 of them for an average of nine years. He identified three groups of patients. About a third were thought to have a hysterical conversion syndrome as well as an organic diagnosis, and on follow up the organic disease prevailed. About a third were initially thought to have pure hysteria, of whom eight later developed an organic disease, and about a third had a psychiatric diagnosis including schizophrenia, depression, and personality disorder. Unfortunately, insufficient details are given to determine whether the subsequent organic disease explained the presenting features, particularly as the eventual diagnoses included trigeminal neuralgia, thoracic outlet syndrome, Takayasu's disease, and cord compression. Three patients went on to develop dementia. Investigations available today would probably have revealed some of these conditions at the first assessment. Twelve deaths occurred, including four suicides, one in a patient with a myopathy and another in a patient with multiple sclerosis. Of the remaining deaths, three were vascular, one due to a brain stem angioma, and one to a glioma. If relevant to the presenting problem most of these conditions would now be diagnosed by magnetic resonance imaging or other relatively non-invasive investigations. Slater did not confine himself to unexplained motor symptoms, so this series cannot be compared directly with that of Crimlisk et al.

Slater concluded that the diagnosis of hysteria was a disguise for ignorance and a fertile source of clinical error. As a result neurologists and psychiatrists became reluctant to diagnose pure hysteria, concerned that they were missing a neurological disease even if some of the neurological features were clearly non-organic. Nevertheless, assuming proper clinical evaluation and negative results on investigation, the chances of a patient developing a neurological disease that might have accounted for the original complaint is very small. Invasive and perhaps dangerous investigations are not appropriate to exclude rare and untreatable conditions; in most patients it is better to recognise the non-organic component of the problem and to seek psychiatric advice. The diagnosis can be reviewed if any new features develop.