A functional MRI study of working memory task in euthymic bipolar disorder: evidence for task-specific dysfunction

• Published in: Bipolar disorders 2004-12, Vol.6 (6), p.550-564
• Main Authors: Monks, Paul J, Thompson, Jill M, Bullmore, Edward T, Suckling, John, Brammer, Michael J, Williams, Steve CR, Simmons, Andrew, Giles, Nicola, Lloyd, Adrian J, Louise Harrison, C, Seal, Marc, Murray, Robin M, Nicol Ferrier, I, Young, Allan H, Curtis, Vivienne A
• Format: Article
• Language: English
• Subjects: bipolar disorder; Bipolar Disorder - drug therapy; Bipolar Disorder - epidemiology; Brain - metabolism; Diagnostic and Statistical Manual of Mental Disorders; Disability Evaluation; Dysthymic Disorder - drug therapy; Dysthymic Disorder - epidemiology; functional magnetic resonance imaging (fMRI); Humans; Lithium Carbonate - therapeutic use; Magnetic Resonance Imaging; Male; Memory Disorders - epidemiology; Memory Disorders - metabolism; Middle Aged; neuropsychological dysfunction; prefrontal cortex; verbal working memory
• ISSN: 1398-5647; 1399-5618
• DOI: 10.1111/j.1399-5618.2004.00147.x

Objectives:  Even when euthymic bipolar disorder patients can have persistent deficits in working memory, but the neural basis of this deficit remains unclear. We undertook an functional magnetic resonance imaging investigation of euthymic bipolar disorder patients performing two working memory paradigms; the two‐back and Sternberg tasks, selected to examine the central executive and the phonological loop respectively. We hypothesized that neuronal dysfunction would be specific to the network underlying the executive rather than the phonological loop component of working memory. Methods:  Twelve right‐handed euthymic bipolar I males receiving lithium carbonate monotherapy were matched with 12 controls. The two‐back task comprised a single working memory load contrasted with baseline vigilance condition. The Sternberg paradigm used a parametric design incorporating variable working memory load with fixed delay between presentation of an array of items to be remembered and a target item. Functional activation data were acquired during performance of the tasks and were analysed to produce brain activation maps representing significant group differences in activation (ANOVA). Load–response curves were derived from the Sternberg task data set. Results:  There were no significant between‐group differences (t‐test) in performance of the two‐back task, or in 2 × 5 group by memory load ANOVA for the performance data from Sternberg task. In the two‐back task, compared with controls bipolar disorder patients showed reductions in bilateral frontal, temporal and parietal activation, and increased activations with the left precentral, right medial frontal and left supramarginal gyri. No between‐group differences were observed in the Sternberg task at any working memory load. Conclusions:  Our findings support the notion that, in euthymic bipolar disorder, failure to engage fronto‐executive function underpins the core neuropsychological deficits.