Monica Åström

---

Depression after stroke

BOTH STROKE AND depression are major health problems in the elderly. In this study, the prevalence of major depression after stroke was investigated in a well-defined sample of acute stroke patients (n=80), followed up at 3 months, 1 year, 2 and 3 years after the stroke event. Links to biological and psychosocial factors were examined. Hypercortisolism was studied by the dexamethasone suppression test and compared with healthy elderly. Living conditions (including demographic caracteristics, economic resources, health, functional ability, activity/leisure, social network) and life satisfaction were described before and after stroke in relation to a general elderly population. Demographic caracteristics, economic resources, social network and psychiatric morbidity prestroke did not differ from the general elderly population. Already prior to the stroke, patients reported more health problems and lower functional ability in many aspects of daily life, more passive leisure time and a lower global life satisfaction. After stroke, contacts with children were maintained, whilst contacts outside the family declined and remained lower than in the general elderly population. Stroke involved a marked reduction in global life satisfaction. Poor life satisfaction at 1 year remained poor for the entire three years; these stroke victims had a higher frequency of major depression early after stroke. The prevalence of major depression was 25% at the acute stage, 31% at 3 months, decreased to 16% at 1 year, was 19% at 2 years and increased to 29% at 3 years. The most important predictors of immediate major depression were left anterior brain lesion, dysphasia, and living alone. Dependence in self-care ability and loss of social contacts outside the family were the most important predictors at 3 months. From 1 year onwards, loss of social contacts contributed most to depression and at 3 years also cerebral atrophy. Sixty percent of patients with early depression (0-3 months) had recovered at 1 year; those not recovered at 1 year had a high risk of chronicitation. Hypercortisolism as measured by the dexamethasone suppression test was associated with major depression late (3 years) but not early (0-3 months) after stroke. At 3 years, the dexamethasone suppression test had a sensitivity of 70%, a specificity of 97%, a positive predictive value of 88%, a negative predicitive value of 91%, and a diagnostic accuracy of 90%. Nonsuppression of dexamethasone at 3 months was a significant predictor of major depression at 3 years.