Cognitive impairment in bipolar disorder: Neurodevelopment or neurodegeneration? An ECNP expert meeting report

Abstract

This is a report arising from an ECNP expert meeting. Recent studies have focussed on cognitive problems in manic-depressive illness and a few have addressed premorbid neuropsychological functioning. The results are not fully consistent but seem to point to a neurodegenerative model, rather than a neurodevelopmental one, for some cognitive domains. There is agreement that cognitive dysfunction is highly correlated with psychosocial functioning. The neurobiological and clinical implications of recent findings will be discussed. Treatments to reduce subsyndromal symptoms and relapses may indirectly improve neurocognitive deficits and this should be better documented. Moreover, neurocognitive impairment in bipolar disorder should be considered a potential therapeutic target, so that research should focus on new drugs and psychological interventions, including neurocognitive rehabilitation, addressed to improve not only the cognition but also the functional outcome of this population.

Introduction

There is a broad consensus that patients with bipolar I and bipolar II disorders have cognitive impairments when, compared with age matched controls of similar educational background (Bearden et al., 2001, Quraishi and Frangou, 2002). Furthermore, these impairments appear to be somewhat independent of affective state, as they are present in euthymic or symptom free individuals (Martinez-Aran et al., 2004b, Robinson et al., 2006). The cognitive domains that appear to be most consistently impaired are: attention, memory and executive function (Quraishi and Frangou, 2002). On average, the magnitudes of the detected effects are between, on average, 0.2 to 1 standard deviation unit (Thompson et al., 2005, Robinson et al., 2006) and they are, broadly speaking, smaller than those seen for cohorts with schizophrenia of a similar age (Martinez-Aran et al., 2002). A recent meta-analysis found moderate to large effect sizes in most neurocognitive measures especially on verbal memory, attention/processing speed and executive functions (Torres et al., 2007). In a two-year follow-up study by Mur et al. (2008), euthymic bipolar patients showed a poorer neurocognitive performance between − 1 and− 2 SD compared to healthy controls on executive function and processing speed measures.

However, it would be wrong to say that there exists an obvious pattern of impairment that specifically characterizes bipolar disorder, although some studies have sought to define differences from schizophrenia or psychosis (Clark et al., 2005b, Glahn et al., 2006, Glahn et al., 2007). However, the cause(s), consequences and potential prevention of cognitive impairments in bipolar disorder are of great contemporary interest. Of particular importance is the nature of these impairments and how they may inform us about the aetiology of bipolar disorder. The primary aim of this review is to determine if the current literature can address the following question. Specifically, are cognitive impairments found in bipolar patients more consistent with neurodevelopmental or neurodegenerative disorders?