Letters

Prognosis of symptoms that are medically unexplained

BMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7157.536 (Published 22 August 1998) Cite this as: BMJ 1998;317:536

Every neurology service should have access to specialist liaison psychiatry

  1. Michael Götz, Consultant psychiatrist,
  2. Allan House, Senior lecturer in psychiatry
  1. Kildean Day Hospital, Stirling FK8 1RW
  2. School of Medicine, Division of Psychiatry and Behavioural Sciences in Relation to Medicine, Leeds LS2 9LT
  3. Section of Psychological Medicine, School of Postgraduate Medical Education, University of Warwick, Coventry CV4 7AL
  4. Newcastle General Hospital, Newcastle upon Tyne NE4 6BE
  5. Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP
  6. Division of Psychosomatic and Liaison Psychiatry, St George's Hospital Medical School, Tooting, London SW17 0RE

    EDITOR—Crimlisk et al's report on the outcome of motor symptoms that were medically unexplained in a cohort first identified at the National Hospital for Nervous Diseases is welcomed by those interested in the psychiatry of physical illness.1Slater's work, based on cohorts recruited in the 1950s, overestimated the likelihood of misdiagnosis in modern neurology,2 and his conclusions have dissuaded psychiatrists from becoming involved in the management of hysteria.

    O'Brien's editorial on Crimlisk et al's study could have sounded two cautionary notes.3Firstly, the study was not based on a true inception cohort, and it was recruited from a highly specialised centre. Usually, these factors lead to the finding of a worse prognosis than if only incident cases were included and the sample was less prone to recruitment bias. In this case, however, the bias may have been in the other direction. Patients seen at the National Hospital are likely to have been more exhaustively investigated and observed for longer (either in this episode or before referral) than they would be in most hospitals, so that the chances of misdiagnosis are reduced. We should not generalise the study findings to hospitals where initial assessment may be undertaken with little or no access to specialist neurological opinion or investigations.

    Secondly, the negative message—that high rates of undetected neurological disease are not seen at follow up—has been allowed to obscure the important positive findings of the study. Even years later, the patients had disabling physical symptoms and substantial psychiatric problems. As in Slater's original study, preventable deaths occurred from suicide and the complications of immobility. Many people were apparently not referred for psychiatric help either at the time of the initial presentation or subsequently.

    The important conclusion of Crimlisk et al's study must be that every neurology service should have easy access to referral to specialist liaison psychiatry.4 This group of patients, who are difficult to treat, are often resistant to treatment, and have a poor prognosis, may then have a reasonable chance of obtaining appropriate treatment for their primary disorder.

    References

    Clinical guidelines are needed

    1. Chris J Mace, Senior lecturer
    1. Kildean Day Hospital, Stirling FK8 1RW
    2. School of Medicine, Division of Psychiatry and Behavioural Sciences in Relation to Medicine, Leeds LS2 9LT
    3. Section of Psychological Medicine, School of Postgraduate Medical Education, University of Warwick, Coventry CV4 7AL
    4. Newcastle General Hospital, Newcastle upon Tyne NE4 6BE
    5. Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP
    6. Division of Psychosomatic and Liaison Psychiatry, St George's Hospital Medical School, Tooting, London SW17 0RE

      EDITOR—Crimlisk et al's study of the prognosis of unexplained motor symptoms1repeats that of Slater 30 years ago.2 In an editorial O'Brien requests that other groups with unexplained symptoms be compared,3but he takes no account of the findings of the only previous follow up study from the National Hospital for Neurology and Neurosurgery comparable in size to Slater's study.4That follow up study looked at the clinical history of patients presenting between 1978 and 1980 with unexplained seizures and sensory deficits as well as motor symptoms, in similar proportions to those in Slater's study. All of the patients had been referred for a psychiatric opinion, unlike either Slater's or Crimlisk et al's patients. Forty one per cent of our patients (10 year follow up) failed to improve compared with 52% of patients in Crimlisk et al's study (six year follow up), with two of our patients going into remission after between six and 10 years. This percentage was maintained across the sample (n=73), which included large subgroups of patients with motor symptoms (n=31) and pseudoseizures (n=27).

      The examination of predictive factors evident on initial examination showed associations between persistence of symptoms and length of history. Psychiatric diagnoses of affective and personality disorders, confirmed by Crimlisk et al for their patients with motor symptoms, could be made irrespective of index symptom. Only three of our patients, belonging to the subgroup of patients with motor symptoms, developed neurological disease that accounted for the initial symptom; organic disease had not been suspected at presentation. The initial diagnosis was probably complicated by the presence of other unexplained symptoms in two of these patients and by a temporary response to behaviour therapy in the third.

      Crimlisk et al's study is methodologically rigorous. Future studies, however, should clarify the concept of organicity that they used in selecting patients and in assessing outcome. A reliance on consensus will restrict the generalisability of findings. Several additional factors that were evident at assessment and that were likely to influence judgments of organicity were independently associated with clinical outcome in the earlier study.4hese factors included the number of neurological signs (irrespective of type), the prescription of non-psychotropic drugs, a provisional neurological diagnosis for a symptom, and a recorded history of psychiatric help.

      Future studies of patients with unexplained symptoms need to collect initial clinical data on individual findings and signs. The validity of clinical data as indicators of organicity can then be established in the light of outcome. If reliable and objective clinical guidelines can then be produced this would benefit all assessors of patients with unexplained symptoms.

      References

      Psychological aspects of investigations must be addressed early

      1. Alan Currie, Consultant psychiatrist,
      2. Ann Ryman, Senior registrar in psychiatry
      1. Kildean Day Hospital, Stirling FK8 1RW
      2. School of Medicine, Division of Psychiatry and Behavioural Sciences in Relation to Medicine, Leeds LS2 9LT
      3. Section of Psychological Medicine, School of Postgraduate Medical Education, University of Warwick, Coventry CV4 7AL
      4. Newcastle General Hospital, Newcastle upon Tyne NE4 6BE
      5. Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP
      6. Division of Psychosomatic and Liaison Psychiatry, St George's Hospital Medical School, Tooting, London SW17 0RE

        EDITOR—In his editorial on neurological symptoms that are medically unexplained O'Brien mentions the dangers of inappropriate investigations and the unprofitability of repeat investigations for the same complaint.1 He fails to point out, however, that such investigations perpetuate an organic view of the origin of the symptoms and make it harder to tell patients later that there may be a psychological component to the aetiology of their symptoms. Investigations aimed at reassuring the patient may have the opposite effect unless psychological aspects are addressed early.2

        Of greater concern, however, is O'Brien's interchangeable use of the terms non-organic symptoms and hysteria. It is inappropriate to label all non-organic symptoms as psychiatric in origin as they are not all associated with clearly discernible psychiatric morbidity. 3 4When an association does exist between non-organic symptoms and psychiatric morbidity this is a medical explanation for the symptoms. To say otherwise artificially separates the practice of psychiatry from the rest of the medical specialties.

        Practitioners should ask for psychiatric advice if investigations are inconclusive, but diagnosis by exclusion and looking for psychiatric morbidity only at this late stage can be problematic. A facility for re-evaluation should primarily be reserved for those patients whose symptoms either progress or are truly medically unexplained.

        References

        Follow up study needs to be continued for longer

        1. John Farnill Morgan, Clinical research fellow
        1. Kildean Day Hospital, Stirling FK8 1RW
        2. School of Medicine, Division of Psychiatry and Behavioural Sciences in Relation to Medicine, Leeds LS2 9LT
        3. Section of Psychological Medicine, School of Postgraduate Medical Education, University of Warwick, Coventry CV4 7AL
        4. Newcastle General Hospital, Newcastle upon Tyne NE4 6BE
        5. Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP
        6. Division of Psychosomatic and Liaison Psychiatry, St George's Hospital Medical School, Tooting, London SW17 0RE

          EDITOR—Crimlisk et al's attempt to “revisit Slater” does not quite live up to its title.1Slater's seminal study described a 10 year follow up of 85 patients referred to the National Hospital for Nervous Disease, of whom about a third developed an established organic illness within 7-11 years.23Crimlisk et al have followed up their subjects after only six years, and it is premature to conclude that “the emergence of a subsequent organic explanation for these [motor] symptoms is rare.”

          Although the authors emphasise that good follow up data were obtained on nine tenths of their patients, only three quarters were re-examined clinically. As 55 patients underwent full interview and examination but 59 patients were assessed with the scale of affective disorders and schizophrenia, four patients assessed with the scale must have been interviewed by telephone. A telephone interview and the examination of medical records are inadequate for establishing diagnoses such as personality disorder or somatisation disorder.

          These criticisms are balanced against the clarity of sampling from a well circumscribed group of patients with unexplained motor symptoms and the pursuit of prognostic indicators. Crimlisk et al's findings, however, are the result of a short term to medium term follow up study with a relatively high attrition rate. If more definitive results are available from the same sample in five years' time the sobriquet “Slater revisited” might truly be deserved.

          References

          View Abstract

          Sign in

          Log in through your institution

          Subscribe