The effect of systemic antidepressant treatments in early stage on neurocognitive function of euthymic bipolar patients initiated with a depressive onset: An observational, cross‐sectional, single‐blind study protocol

Bipolar disorder (BD) is a severe chronic disorder characterized by depressive episodes and alternating or intertwining manic and hypomanic episodes, with a lifetime prevalence rate exceeding 1% worldwide (Clemente et al., 2015). BD can result in a wide range of neurocognitive dysfunctions, mainly involving attention, verbal learning, nonverbal memory, and executive functions (Douglas et al., 2018; Kurtz & Gerraty, 2009). The neurocognitive dysfunctions could occur in very early stage of illness and persist even in euthymic period (Sole et al., 2017), which causes severe damage to individual psychosocial function and significantly increases disease‐related burden (Alonso et al., 2011). Therefore, alleviating neurocognitive dysfunction and accordingly improving patients’ psychosocial function and quality of life have become a critical goal of BD management (Grande et al., 2017). To this end, some clinical studies have attempted to improve the neurocognitive function of patients with BD pharmacologically and psychotherapeutically, but the outcomes are far from satisfaction (Sole et al., 2017). One of the reasons could be that the understanding on the risk factors related to neurocognitive impairment in such group of people is limited. Theoretically, a comprehensive and in‐depth understanding of relevant risk factors will help to better guide the prevention and treatment work of BD.

About 50% of BD initiates its expression in the form of depressive episode (Drancourt et al., 2013; Tondo et al., 2010), in which only 20% can get a diagnosis of BD within the first year of onset (Hirschfeld et al., 2003), and the average delay from first episode to diagnosis of BD is 5–10 years (Baldessarini et al., 2007). The main reason for this significant delay is that a large number of bipolar patients do not experience any manic/hypomanic symptoms in the early stage, defined as a term from initial depressive onset to the diagnosis of BD, and the clinical manifestations of bipolar depression are highly similar to unipolar depressive episodes (Drancourt et al., 2013; Tondo et al., 2010). As a result, patients may receive systemic treatment with antidepressants for a long time because of the misdiagnosis or delayed diagnosis. The attitude regarding the effect of antidepressants used in early stage on neurocognitive function of patients with BD is controversial yet. Empirically, antidepressant treatment could deprave the natural course of disease and facilitate switching from depressive onset to episodes of mania, hypomania, or mixed state (Berkol et al., 2019; Sole et al., 2017; Viktorin et al., 2014), which may cause cognitive injury (Goodwin et al., 2008). For example, BD patients with higher number of manic and/or hypomanic episodes had a worse performance in Neurocognitive Composite Index (NCI), visual memory, and working memory (Sánchez‐Morla et al., 2019); furthermore, patients with BD, particularly those with multiple mood episodes, showed severe dysfunction in emotional memory than healthy controls (Fijtman et al., 2020). In addition, the number of manic episodes was positively correlated with level of 8‐hydroxy‐2 deoxyguanosine (8‐OHdG), a modulator of DNA methylation, which could be a potential mechanism underlying cognitive dysfunction (Soeiro‐de‐Souza et al., 2013). Consistently, studies have shown that antidepressant agents per se undermined the neurocognitive function of the elderly and increased the incidence of mild cognitive impairment or dementia (Leng et al., 2018; Moraros et al., 2017), which may contribute partly to the neurocognitive impairments in patients with BD. However, it is arguable to expand the view based on the evidence from bipolar depression to depressive episode prior to any mania/hypomania emergence. Furthermore, it has been illustrated that antidepressant treatment can improve cognitive performance (Brendel et al., 2018; Tunc‐Ozcan et al., 2019), although the evidence did not derive from studies focusing on bipolar patients specifically. In view of the high incidence of delayed diagnosis or misdiagnosis of BD (Tondo et al., 2010), the use of antidepressants in the early stage is quite common and extensive. Therefore, a further understanding in the relationship between antidepressant used in the early stage and neurocognitive function of bipolar patients has significantly scientific and clinical significance.

The current study attempts to clarify this question by making comparisons in neurocognitive function between bipolar patients who had a depressive episode as initial onset and being with and without systemic antidepressant treatment (received adequate dosage of antidepressants for 6 weeks at least) in early stage. Furthermore, the social functions and quality of life will also be investigated because both items are highly associated with cognitive functions and are indicative for functional rehabilitation. Apart from those, we will also study the influence of antidepressants used in early stage on impulsiveness, quality of sleep, and positive and negative affect. Lastly, peripheral neuron‐specific enolase (NSE) will be evaluated. NSE, a cell‐specific isoenzyme of glycolytic enolase, is known as a sensitive marker of neuronal damage (Isgrò et al., 2015). It has been showed that high level of serum NSE was related to mild cognitive dysfunction in patients with diabetic retinopathy (Yu et al., 2020) and that serum NSE levels of BD patients were significantly different from healthy controls (Akcan et al., 2018). In this regard, it is expected to examine the potential association between serum NSE level and cognitive function of BD patients. Finally, although it is mainly an explorative investigation and is informative for clinical practice regardless of outcomes, we expect that systemic antidepressant treatments in early stage could negatively affect the neurocognitive function, social functions, and quality of life of patients with BD.

Regarding the critical role of cognitive impairment in psychosocial disability and poorer quality of life, improving cognitive function has become a crucial goal in the management of BD patients. In this regard, increasing amount of research examining pharmacological and psychotherapeutic treatments was performed during the past decade (Alda et al., 2017; Solé et al., 2015). The research outcomes are inconsistent and no therapeutic intervention with sufficient efficacy is well‐established for improving cognitive dysfunction of BD yet (Miskowiak et al., 2016), although some of psychotherapeutic treatments or drugs appear as promising candidates. Many factors contributed to the discrepant outcomes such as poor randomization procedure, age of onset, number of depressive and manic episodes, severity of symptoms, and medications used specifically. In view of significantly delayed diagnosis of BD and common use of antidepressants in early stage, the effect of antidepressant agents on cognitive function has attracted increasing attention. But the data are still relatively rare, and the evidence is controversial. For example, a meta‐analysis estimated the relationship between antidepressant use and cognitive impairment and found that the odds of cognitive impairment were positively associated with antidepressant use in participants with an average age equal to or more than 65 years (odds ratio [OR] = 1.65), whereas those with participants aged less than 65 revealed an even stronger association (OR = 3.25) (Moraros et al., 2017); and also the study by Leng and colleagues illustrated that users of selective serotonin reuptake inhibitors and trazodone were as more as two to three times to develop dementia compared with the nonusers 5 years later, although the use of tricyclic antidepressants or other antidepressants was not involved in cognitive impairment (Leng et al., 2018). The evidence above suggests that antidepressant use could be a contributor for cognitive impairment of patients with mood disorder. Importantly, it is believed empirically that antidepressant treatment could deprave the natural course of BD and switch from depressive episode to episode of mania, hypomania, or mixed state (Viktorin et al., 2014), which may undermine the cognitive performance of bipolar patients (Goodwin et al., 2008). For instance, it was found that the higher number of manic and/or hypomanic episodes was negatively correlated with the cognitive performance of euthymic BD patients (n = 99) such as NCI, visual memory, and working memory in a 5‐year follow‐up longitudinal study (Sánchez‐Morla et al., 2019); furthermore, BD patients (n = 33), especially those with multiple mood episodes including mania, hypomania, and/or mixed state, showed severe impairments in emotional memory than healthy controls (n = 20) (Fijtman et al., 2020). Considering the significantly delayed diagnosis of BD and common use of antidepressants in clinical practice, such potential scathing effects should gather more attention. In contrast, another meta‐analysis suggested that antidepressants played a significant positive role in promoting psychomotor speed and delayed recall, with no significant difference between diverse types of antidepressants (Rosenblat et al., 2015). This is in line with the evidence from preclinical research that antidepressant agents could promote executive function and environmental information processing through integrating newborn neural cells into the hippocampal dentate gyrus (Tunc‐Ozcan et al., 2019). However, these data rarely came from clinical trial specifically focusing on the use of antidepressants in bipolar patients in early stage. To this end, the current study aims to clarify this issue by examining the cognitive function between bipolar patients who initiated with depressive episode and being with and without systemic antidepressant treatments in early stage. Additionally, the contemporary management of disease not only requires symptomatic relief but also pursues functional rehabilitation. Therefore, we will also examine the effect of antidepressant treatments in early stage on global function and quality of life of participants because both aspects are highly associated with cognitive function. Considering the potential influence of antidepressant use on progressing of course, the information on sleep quality and positive and negative affect will also be collected. Finally, the peripheral level of NSE will be tested to explore the potential biological mechanism of cognitive impairment.

To increase the power of this study, we have some considerations as following. First, we will perform the PAS measurement to adjust the potential difference in premorbid function that may confound the cognitive performance of participants. Second, it is required that participants should have a euthymic state for 4 weeks at least before enrollment because the duration in euthymia is associated with the quality of life (Pascual‐Sánchez et al., 2019), which has been found associated to cognitive function (Sánchez‐Morla et al., 2009). Third, we have two separate groups of investigators before and after group assignment to keep the later investigators blind to the assignment and then to avoid subjective bias on assessment of scales. Lastly, we set the range of age between 18 and 50 years to decrease the influence of aging in cognitive performance.

In summary, despite these limitations, the present study is expected to help us look further into the effect of antidepressant use in early stage on the cognitive function of patient with BD and may provide clues for inspecting the pathophysiological association/discrepancy between unipolar depression and bipolar depression.

The authors declare no conflict of interest.

HY and CY conceived the study. YL and HY drafted the first version of manuscript. YL and XL provided the methodological supports. CY acquired funding. CY and XL finished the review and editing of protocol. All authors provided input and read and approved the final version.