The role of psychological factors and lesion location in early onset post-stroke depression

Stroke is associated with higher than average rates of depression, with approximately one third of survivors affected (Hackett, Yapa, Parag, & Anderson, 2005). Despite its high prevalence, clinically significant depression following stroke, commonly referred to as post-stroke depression, is often undetected and subsequently not treated. The possible consequences of post-stroke depression include poorer rehabilitation outcomes, cognitive impairment and increased mortality. An understanding of the most appropriate interventions for post-stroke depression is lacking. The aims of this research were to examine factors associated with post-stroke depression that have the potential to guide future pharmacological or psychosocial treatment or preventive interventions. This objective was met by exploring the relationship between depression within the early post-stroke period and: (i) lesion location on MRI using a data-driven approach; (ii) self-concept; and (iii) dimensions of perceived availability of social support. First-episode stroke patients who presented to Monash Medical Centre and Dandenong Hospital between May 2009 and September 2010 were considered for inclusion in this study (approximately 650 patients). Seventy-one patients were recruited with 60 taking part in all aspects of the study. Patients were assessed within 12 days (acute assessment) and at one month (follow-up assessment) post-stroke. Infarcts identified on magnetic resonance imaging (MRI) scans were analysed. Infarct patterns in stroke patients with and without depression were compared using statistical parametric mapping. Self-concept was measured using the validated Tennessee Self-Concept Scale (TSCS:II) and the Head Injury Semantic Differential (HISD-III). The HISD-III measures pre-stroke self-concept retrospectively. Dimensions of perceived social support were measured using the Medical Outcomes Study Social Support Inventory (MOS). Patients with an a priori defined cut-off score of 11 on the Hospital Anxiety and Depression Scale at follow-up were further assessed using the Mini-International Neuropsychiatric Interview (DSM-IV criteria). Multivariate logistic regression analyses were used to assess the independent contribution of self-concept and social support to depression. Twenty-seven percent of the sample experienced depression; 15% major and 12% minor. Results of the voxel-based analysis were inconclusive; however this was influenced by minimal overlap of infarct location between patient scans, thus implicating the interpretation of results. Additionally, the imaging study did not provide evidence for the role of lesion location in post-stroke depression. In relation to self-concept, HISD-III ratings were significantly decreased post-stroke compared to retrospectively-reported pre-stroke ratings (p < 0.05). Results also indicated that self-concept declines in the acute post-stroke period and a more negative self-concept is independently associated with subsequent post-stroke depression. In relation to social support, depressed stroke patients perceived lower availability of emotional/information support and positive social interaction at follow-up compared to non-depressed patients, whereas other domains were not significantly related. In conclusion, this study investigated novel psychosocial variables in relation to post-stroke depression and applied more recent neuroimaging technology to examine the role of lesion location. The implications of this research include the unveiling of psychosocial variables that are amenable to change and thus could be targeted in future interventions for post-stroke depression. Additionally, the results of the lesion location study have provided useful information and several methodological considerations were proposed for future investigations.