Background: Working memory as a part of higher-order executive functions is defined by the parallel storage and processing of information. Recent functional fMRI studies have revealed a functional, interregional disintegration of a neuronal network connecting cortical, subcortical and cerebellar regions in schizophrenic patients (SZ). Cognitive impairment in working memory is a core psychopathological correlate of schizophrenic symptoms. Atypical neuroleptics such as quetiapine have shown good efficacy in treating positive and negative symptoms. The presented study evaluated the impact of a neuroleptic steady state treatment with quetiapine on the altered working memory activation patterns in schizophrenia.
Methods: Patients were examined by fMRI at baseline and after 12 weeks of steady state treatment with quetiapine. Matched healthy controls (HC) underwent baseline examination. In the scanner, stimuli were presented in a 2-back and 0-back condition of a working memory (wm) paradigm, whereby a degraded and a non-degraded version were used each time. Additionally, behavioural responses (reaction time to target stimuli and error ratio) were measured.
Results: At baseline, healthy controls revealed increased activity in the frontal lobe, especially in regions of the prefrontal cortex. Compared to HC, SZ showed hypoactivation in the right dorsolateral prefrontal cortex (DLPFC) and the ventrolateral prefrontal cortex (VLPFC) bilaterally for the 2-back condition. In the 2-back degraded condition there was a hypoactivation in both, the right DLPFC and the VLPFC. Additionally, patients showed bilaterally decreased activation in the basalganglia in the 2-back and in the right caudatus in the 2-back degraded condition compared to healthy controls. After treatment with quetiapine, patients activations patterns were increased. The pre-post comparison of the 2-back condition revealed a significant increase of activation in the left VLPFC at a significance level of 0.001 (uncorrected). The 2-back degraded condition led to a significant activation pattern in the lingual gyrus and the right precuneus. In both wm conditions, at baseline there were no differences in reaction time but only a worse performance in SZ. After treatment, behavioural measurement of responses, including reaction time and performance, showed slight improvements in SZ, although these did not reach statistical significance.
Conclusions: The neuronal networks underlying working memory are clearly altered in schizophrenia. After 12 weeks of treatment with quetiapine monotherapy, patients showed significant clinical improvement and revealed increased BOLD activity in the VLPFC during a working memory task, although there was no improvement of cognitive performance.