The utility of the dexamethasone suppression test (DST) in the diagnosis of depression were examined in 38 stroke inpatients. The depressive symptoms were assessed by the modified Hamilton depression rating scales.

The results were as follows:

1) The patients with abnormal DST had significantly higher Hamilton scales (13∼15) than those with normal DST (8∼9).

2) Defining the depression as above 14 points of Hamilton scales, the sensitivity, specificity and positive predictive value (PPV) of the DST were in the following order: 69.2∼78.6%, 79.2∼87.5% and 66.7∼76.9%.

3) Three patients with false positive DST were all within one month after the stroke,

4) The specificity and PPV of the DST were in the following order: DSTL (DST in which the lower cortisol value between the 8AM and 4PM cortisol, value was selected)- 87.5%, 76.9%; DST8AM (DST in which the 8AM and 4AM cortisol value was selected(- 87.0%, 75.0%; DST4AM (DST in which the 4AM cortisol value was selected)- 82.6%, 66.7%.

5) Abnormal DST or Hamilton scales representing depression were not correlated with the nature, site or size of the brain lesion.

The results demonstrate that the DST is a useful specific indicator of depression in patients with stroke, and it is necessary to rule out the false positivity, especially in acute phase.