Background: Poststroke depression (PSD) is a common and invalidating condition, requiring a prompt diagnosis to provide appropriate treatment. The diagnosis of PSD is based on clinical evaluation, supported by psychiatric scales, predominantly constructed to assess 'functional' depressive disorders, which can miss important clinical information about PSD. This study evaluated the diagnostic accuracy of the Poststroke Depression Rating Scale (PSDRS), a diagnostic tool specifically devised to assess depression after stroke, in comparison to the Hamilton Depression Rating Scale (Ham-D). Methods: 143 patients were enrolled at their first-ever stroke; 46 subjects received a diagnosis of major depression-like disorder (MDL) and 53 a diagnosis of mood disorder with depressive manifestations (MDDM). Each patient underwent the PSDRS, Ham-D and Mini Mental State Examination. Areas under receiver-operating characteristic curves were calculated to evaluate diagnostic accuracy. Results: At their optimum cut-off points, the Ham-D and PSDRS showed good sensitivity and specificity for MDL or MDL + MDDM; the PSDRS had a higher positive predictive value for MDL in respect of the Ham-D (78 vs. 59%). Furthermore, the diagnostic accuracy of the PSDRS was higher in respect of the Ham-D in aphasic patients. Regression analyses showed that a longer latency from stroke onset predicted misdiagnosis on both PSDRS and Ham-D; the Ham-D was significantly influenced by cognitive impairment while the PSDRS was affected by age. Conclusions: The Ham-D and PSDRS are both reliable diagnostic tools for the diagnosis of PSD. The PSDRS showed a slightly superior positive predictive value for MDL, particularly in aphasic patients, and was not influenced by cognitive dysfunction, a common consequence of stroke affecting the diagnostic performance of the Ham-D. We suggest that the PSDRS could be a useful tool in clinical practice and in therapeutic trials.
- Hamilton depression rating scale
- Mood disorders
- Poststroke depression rating scale
ASJC Scopus subject areas
- Clinical Neurology