The last decade has witnessed unprecedented scientific progress in the area of bipolar disorder. This is shown by the fact that there has been an exponential growth of publications (1) and of new drug indications. Nowadays, it is not rare to see journal issues like the present one, where articles on bipolar disorder represent a great deal of what is being published (2–7), even in journals like Acta Psychiatrica Scandinavica, which deal with all the range of mental disorders. One of the most recent and productive areas of research in the field of bipolar disorder has been neurocognition, as one can easily track by the number of citations of the publications that have focused on that topic. Bipolar disorder is no longer free of cognitive impairment, as assumed in the past, but actually a condition in which subtle and mild impairments persist over remission and have an outstanding impact on psychosocial functioning (8–10). In the latest years we have been able to disentangle the factors that lead to neurocognitive impairment in bipolar disorder, including genetics (11, 12), subthreshold depressive symptoms (13, 14), the number of episodes (15), particularly the manic ones (16), a history of psychotic features (17), substance abuse (18), poor adherence (19), and of course, medication (20, 21). Medication is actually a two-edge sword because on one hand it improves cognition by targeting psychotic and mood symptoms, but on the other hand it carries its own cognitive side-effects. A traditional example is lithium (22), which can be neuroprotective as well as neurotoxic (23), but in fact no psychotropic drug is free of neuropsychological side-effects. Potential mechanisms underlying those include anticholinergic, sedative, extrapyramidal, and blunting effects, which may vary from drug to drug. Although some drugs appear better than other with this regard (24), large and fully powered randomized comparative trials are not available as yet.

In this issue of Acta Psychiatrica Scandinavica, Goswami et al. (4) report that a sample of 22 drug-free euthymic bipolar patients had similar neuropsychological performance than a comparable sample of 22 medicated patients. Their finding is extremely relevant because medication has traditionally been the main confounder in studies comparing patients with bipolar disorder and healthy subjects. Although it could be argued that sample size was far from ideal, there were more high achievers in the medicated group, and that no subjects received antipsychotics, anticholinergics, anxiolytics or hypnotics, which is unusual and might explain their relatively good performance, the fact is that differences were small, suggesting that the effects of medication may be negligible. Ideally, a healthy, unmedicated group might have confirmed that differences between groups would be basically related to illness-related features. The relative unimportance of medication on bipolar disorder-associated deficits indicates that one of the possible ways to treat and prevent the neurocognitive impact of recurrences could be early, intensive, continued pharmacological treatment, addressed at tackling subthreshold depressive symptoms and the prevention of further episodes. Other potentially useful tools include psychoeducation (25, 26) and interventions aimed at reverting the cycle of allostatic load (27).

In the near future, specific drugs and psychosocial interventions should aim at targeting neurocognitive symptoms and their functional correlates. Those trials should enroll patients in remission with significant cognitive impairment and prove that the experimental therapy may have positive effects on those deficits and on patient’s functioning, as measured with disability scales such as the functioning assessment short test (FAST) (28). Several drugs, including galantamine, memantine, armodafinil, mifepristone, n-acetyl cysteine, folic acid, and many others are currently under test, as well as neurocognitive remediation techniques, which give us hope that the day when we may achieve true full recovery of bipolar disorder is not too far ahead in time.


  1. Top of page
  2. References