Intended for healthcare professionals


Multiple sclerosis, depression, and suicide

BMJ 1997; 315 doi: (Published 20 September 1997) Cite this as: BMJ 1997;315:691

Clinicians should pay more attention to psychopathology

  1. Anthony Feinstein, Assistant professora
  1. a University of Toronto, Department of Psychiatry, Sunnybrook Hospital, Room FG38, North Tork, Ontario M4N 3M5, Canada

Mental illness leaves patients at risk for harming or killing themselves, none more so than major depression, with which a 15% lifetime prevalence of suicide has been consistently noted.1 Less clear is how these figures translate when applied to patients with neurological disease, particularly those conditions known to be associated with a high risk of comorbid depression.

An example is multiple sclerosis, the leading neurological cause of disability in young and middle aged adults. Depressive symptoms of sufficient severity and duration to warrant a diagnosis of major depression affect up to half of patients during the course of their illness.2 This is three times the prevalence reported for major depression and psychiatric comorbidity in community based samples, and it also exceeds that for other neurological disorders.3 Depression in multiple sclerosis is not linked to a family history of affective disorder,2 nor is it more likely to occur before the start of neurological symptoms.4 While detection of brain lesions by magnetic resonance imaging has been shown to correlate with cognitive dysfunction in multiple sclerosis, a correlation with depression has proved more elusive, with evidence for and against being noted. Evidence that increased social stressors and inadequate family and community support are important suggests that depression in multiple sclerosis has a complex, multifactorial pathogenesis.5

While depression undoubtedly adds to morbidity, some data suggest it may also lead to suicide. Not all studies agree on this, although those that failed to find an association have been the most methodologically flawed, containing small sample sizes and failing to standardise results for age and sex with control populations. One study that overcame these limitations investigated cause of death in a Danish cohort of 5525 patients with onset of multiple sclerosis between 1953 and 1985. Twice the expected number of suicides was found, with men and those with symptoms starting before the age of 30 years being most at risk.6 While of concern, these results should be interpreted with caution as the actual number of patients who killed themselves was small (53), and case registries may not always record the cause of death accurately.

Nevertheless, these findings are supported by two other sources: a Medline search of all published data, which shows that patients with multiple sclerosis were generally more likely to attempt or commit suicide than patients with other common neurological disorders,7 and a Canadian study of 3126 patients with multiple sclerosis who were followed longitudinally at two clinics between 1972 and 1988.8 Comprehensive databases kept track of virtually all patients within the respective catchment areas, each of whom received at least a yearly follow up examination. Suicide accounted for 15% of all ascertained deaths during this 16 year period, proportionately 7.5 times that for the general population matched for age but not sex. While this figure exceeds that of the Danish epidemiological investigation,6 failure to control for sex in a disorder with a known female preponderance should again prompt caution when interpreting the findings. The overall picture to emerge is therefore one of a slightly increased risk of suicide probably related to the high prevalence of depression, although none of the studies cited specifically addressed aetiology.

With these statistics, it is surprising that the treatment of depression in multiple sclerosis has received scant attention. There is only one report of a double blind, placebo controlled drug trial showing that desipramine, although modestly effective, produced troubling anticholinergic side effects that limited the dose.9 Anecdotal reports of fluoxetine being effective and better tolerated suggest that newer antidepressant compounds are probably the treatment of choice. Psychotherapy is a useful adjunct and should not be overlooked. For the acutely suicidal, depressed patient, electroconvulsive therapy remains an option. It may, however, trigger an exacerbation of multiple sclerosis, the presence of contrast-enhancing brain lesions on a pretreatment magnetic resonance scan being a warning sign.10

While the past decade has undoubtedly brought a greater awareness of the neurobehavioural sequelae of multiple sclerosis, the risk remains that clinicians may yet miss a treatable cause of morbidity and mortality. The measure by which disability is assessed remains the expanded disability status scale,11 which affords little weight to psychopathology. Attention remains largely focused on more easily discernable and quantifiable evidence of disease, such as how far patients can walk unaided, the degree of cerebellar disturbance, or measurements of visual acuity. The paucity of studies of treating depression related to multiple sclerosis attests to this. The problem is compounded by new treatment modalities such as interferon beta-1b, in which physical improvement may be offset by a potentially deleterious effect on mood.12

While it is premature to conclude that depressed mood represents a core symptom of multiple sclerosis, it has taken psychiatrists and neurologists almost a century to realise that Charcot's astute observation of altered affect in the disorder he helped define demands prompt and careful management.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
View Abstract