A critical overview of the clinical evidence supporting the concept of neuroprogression in bipolar disorder

Older age bipolar disorder

PURPOSE OF REVIEW

Older age bipolar disorder (OABD) refers to patients with bipolar disorder aged 50 years and over. There is a paucity of evidence-based guidelines specific to OABD, but in recent years, several studies have been published on OABD. The current review synthesizes previous literature (up to January 1, 2021) as well as most recent literature on OABD (since January 1, 2021).

RECENT FINDINGS

This review covers the following themes: diagnosis and specifiers, clinical course, psychosocial functioning, cognition, physical comorbidities, and pharmacotherapy. On the basis of the latest data, specific clinical recommendations are proposed for each theme.

SUMMARY

OABD forms a more complex subgroup of bipolar disorder, with an increased risk of cognitive deficits, physical comorbidities, impaired psychosocial functioning, and premature death. The distinctions between BD-I and BD-II and between EOBD and LOBD do not clinically represent relevant subtypes for OABD patients. Mental healthcare professionals should treat all OABD patients with an integrative care model that takes into account cognitive and physical comorbidities and that contains elements aimed at improvement of psychosocial functioning and quality of life. Older age itself should not be a reason to withhold lithium treatment. Future research should collect data on essential data domains using validated measurement scales.

Mania-related effects on structural brain changes in bipolar disorder - a narrative review of the evidence

Cross-sectional neuroimaging studies show that bipolar disorder is associated with structural brain abnormalities, predominantly observed in prefrontal and temporal cortex, cingulate gyrus, and subcortical regions. However, longitudinal studies are needed to elucidate whether these abnormalities presage disease onset or are consequences of disease processes, and to identify potential contributing factors. Here, we narratively review and summarize longitudinal structural magnetic resonance imaging studies that relate imaging outcomes to manic episodes. First, we conclude that longitudinal brain imaging studies suggest an association of bipolar disorder with aberrant brain changes, including both deviant decreases and increases in morphometric measures. Second, we conclude that manic episodes have been related to accelerated cortical volume and thickness decreases, with the most consistent findings occurring in prefrontal brain areas. Importantly, evidence also suggests that in contrast to healthy controls, who in general show age-related cortical decline, brain metrics remain stable or increase during euthymic periods in bipolar disorder patients, potentially reflecting structural recovering mechanisms. The findings stress the importance of preventing manic episodes. We further propose a model of prefrontal cortical trajectories in relation to the occurrence of manic episodes. Finally, we discuss potential mechanisms at play, remaining limitations, and future directions.

"Apples and pears are similar, but still different things." Bipolar disorder and schizophrenia- discrete disorders or just dimensions ?

Starting with the dichotomous view of Kraepelin, schizophrenia and bipolar disorder have traditionally been considered as separate entities. More recent, this taxonomic view of illnesses has been challenged and a continuum psychosis has been postulated based on genetic and neurobiological findings suggestive of a large overlap between disorders. In this paper we will review clinical and experimental data from genetics, morphology, phenomenology and illness progression demonstrating what makes schizophrenia and bipolar disorder different conditions, challenging the idea of the obsolescence of the categorical approach. However, perhaps it is also time to move beyond DSM and search for more refined clinical descriptions that could uncover clinical invariants matching better with molecular data. In the future, computational psychiatry employing artificial intelligence and machine learning might provide us a tool to overcome the gap between clinical descriptions (phenomenology) and neurobiology.

Does cognitive function in older bipolar patients depend on recurrent or current mood symptoms?

BACKGROUND

Cognitive impairment in patients with bipolar disorder (BD) is viewed as an integral part of the disorder that seems to be rather stable and even present in euthymic state. Current mood symptoms influence cognition negatively and multiple mood episodes could lead to more severe psychopathology and cognitive impairment, resulting in a hypothesized neuroprogressive course of BD. The influence of current mood symptoms and recurrent mood episodes on cognitive functioning warrants further exploration.

METHODS

Cohort 1 included 20 hypomanic and 21 depressed older adults with bipolar disorder (OABD) of which 20 were reassessed in the euthymic state and 50 healthy subjects. Cohort 2 included 27 OABD who had no recurrent mood episodes during 5 years and 29 who had recurrent mood episodes during 5 years. Neuropsychological examination including tests for memory, executive function, attention and verbal fluency was performed repeatedly in all subjects.

RESULTS

In cohort 1 cross-sectional analyses showed that hypomanic, depressed and euthymic patients groups did not differ from each other with respect to their cognitive functioning, except for attention, which was poorer only in depressed patients. Regardless of mood state patients experienced significantly worse cognitive functioning compared to the healthy subjects. Within subject comparisons revealed that performance on memory tasks was worse in patients with current mood symptoms; depressed patients were more impaired in the learning condition and hypomanic patients were more impaired in delayed recall. In cohort 2 cognitive functioning was not different in patients with or without recurrence in 5 year follow-up.

CONCLUSIONS

Although OABD had worse cognitive functioning than healthy subjects, there was a quite stable pattern of cognitive impairment, regardless of current or recurrent mood episodes. These results do not provide consistent support for the hypothesis of neuroprogression in BD.

Longitudinal analysis of cognitive performances in recent-onset and late-life Bipolar Disorder: A systematic review and meta-analysis

OBJECTIVES

Neurocognitive deficits have been widely reported in euthymic Bipolar Disorder (BD) patients and contribute to functional disability. However, the longitudinal trajectory of these deficits remains a subject of debate. Although most research to this date shows that neurocognitive deficits tend to be stable among middle-age BD patients, it remains plausible that deterioration occurs at either early or late stages of this condition.

METHODS

We conducted a comprehensive meta-analysis of studies that reported longitudinal neurocognitive performance among individuals with BD either within the year of their diagnosis or among late-life BD patients. Pooled effects of standardized mean differences (SMDs) for changes in neuropsychological scores over follow-up were estimated using random effects model. We also examined effect moderators, such as length of follow-up, mood state, or pharmacological load.

RESULTS

Eight studies met inclusion criteria for recent-onset and four studies for late-life BD analysis. No evidence for a deterioration in neurocognitive functioning was observed among recent-onset BD patients (8 studies, 284 patients, SMD: 0.12, 95% CI -0.06 to 0.30, mean follow-up: 17 months) nor for late-life BD patients (4 studies, 153 patients, SMD: -0.35, 95% CI -0.84 to 0.15, mean follow-up: 33 months). None of the moderators were shown to be significant.

CONCLUSIONS

These results, when appraised together with the findings in middle-life BD patients and individuals at genetic risk for BD, suggest that neurodevelopmental factors might play a significant role in cognitive deficits in BD and do not support the notion of progressive cognitive decline in most patients with BD.

Three-year longitudinal cognitive functioning in patients recently diagnosed with bipolar disorder

OBJECTIVE

The longitudinal course of neuropsychological functioning after the first manic episode in bipolar disorder is unknown. The present study evaluated cognitive change in bipolar disorder in the first 3 years after the initial manic episode.

METHODS

Ninety-one newly diagnosed patients with bipolar disorder and 61 demographically similar healthy participants received a neuropsychological evaluation assessing multiple cognitive domains at baseline, 1-year, and 3-year time points. Patients also received clinical assessments including mood ratings at all time points.

RESULTS

Patients showed deficits in all domains at baseline, but similar longitudinal trajectories across time relative to healthy participants in most cognitive domains. For processing speed, patients showed more gains than controls from baseline to 1 year, but these gains stabilized thereafter. Patients with alcohol/substance abuse showed an initial delay but subsequent recovery in executive functioning. Patients who discontinued antipsychotic treatment showed better cognitive outcomes in verbal memory.

CONCLUSION

Appropriately treated patients with bipolar disorder showed favorable cognitive outcome in the first 3 years after experiencing an initial manic episode, arguing against cognitive neuroprogression at this stage of the illness. Discontinuation of antipsychotic treatment may be associated with better cognitive outcomes, but clarification of the role of antipsychotics on cognitive functioning requires further investigation.

Structural and Functional Brain Correlates of Neuroprogression in Bipolar Disorder

Neuroprogression is associated with structural and functional brain changes that occur in parallel with cognitive and functioning impairments. There is substantial evidence showing early white matter changes, as well as trajectory-related gray matter alterations. Several structures, including prefrontal, parietal, temporal cortex, and limbic structures, seem to be altered over the course of bipolar disorder, especially associated with the number of episodes and length of the disease. An important limitation is that most of the studies used either a cross-sectional design or a short follow-up period, which may be insufficient to identify all neuroprogressive changes over time. In addition, the heterogeneity of patients with bipolar disorder is another challenge to determine which subjects will have a more pernicious trajectory. Larger studies and the use of new techniques, such as machine learning, may help to enable more discoveries and evidence on the role of neuroprogression in BD.

High BDNF serum levels are associated to good cognitive functioning in bipolar disorder

Background:

Neurotrophins such as brain-derived neurotrophic factor (BDNF), inflammation and oxidative damage may contribute to the pathophysiology of bipolar disorder (BD) in terms of illness activity. To date, there is a lack of studies linking the cognitive impairment observed in BD with these neurobiological mechanisms. This study aimed to investigate the role of these neurobiological factors in clinical and cognitive outcomes in a sample of bipolar individuals.

Methods:

We measured serum BDNF, cytokines and oxidative stress markers in a sample of 133 individuals: 52 euthymic bipolar patients, 32 manic patients and 49 healthy controls. They were all assessed with a comprehensive cognitive battery. Sociodemographic and clinical data were collected. Multiple linear regression models were built to study associations of neurotrophins and inflammatory and oxidative measures with cognitive functioning.

Results:

BDNF levels were decreased in euthymic (p = 0.039) and manic (p < 0.001) individuals. Conversely, inflammatory (interleukin 6 (IL-6)) (p = 0.019) and oxidative stress (p = 0.003) measures were increased in bipolar individuals compared to controls. BDNF levels were associated with executive functioning (β = 0.01, p = 0.02) and verbal memory (β = 0.013, p = 0.005), together with other demographic variables. In particular, verbal memory was also associated with obesity (β=-0.04, p = 0.005). Neither inflammatory markers, oxidative stress markers nor other relevant clinical variables showed any association with cognitive outcome.

Conclusions:

Of all the peripheral neurobiological factors analysed, BDNF was the only one significantly associated with cognitive dysfunction in bipolar disorder individuals. This study emphasizes the role of BDNF not only across mood phases but also in cognitive functioning.

Role of glia in prefrontal white matter abnormalities in first episode psychosis or mania detected by diffusion tensor spectroscopy

BACKGROUND

White matter (WM) abnormalities are amongst the most commonly described neuroimaging findings in patients with psychotic disorders including schizophrenia (SZ) and bipolar disorder (BD), and may be central to pathophysiology. Few studies have directly compared WM abnormalities in patients with SZ and BD in the first episode of illness, and no studies to date have attempted to separate abnormalities of axon and myelin using complementary MRI techniques.

METHODS

We examined WM abnormalities in young adults with SZ (n = 19) or BD (n = 16) within the first year of illness onset, and healthy controls (n = 22) using a combination of diffusion tensor spectroscopy to measure NAA, creatine (Cr), and choline (Cho), and magnetization transfer ratio (MTR). MTR reflects myelin content, NAA diffusion is neuron specific, and Cr and Cho diffusion reflect both neuron and glial signal.

RESULTS

We found no differences in MTR or NAA ADC in either patient group compared to controls, but significant elevations of both Cr and Cho diffusion in patients with SZ, and elevations of Cho diffusion in patients with BD. Elevations in Cr and Cho diffusion in the absence of NAA diffusion abnormalities indicate that the aberrant signal arises in glia.

CONCLUSIONS

Glial abnormalities were present and detectable by the first episode of psychosis, whereas major abnormalities in axon and myelin were not. Examination of these neurobiological markers early in the course of illness may clarify the neuroprogressive nature of these distinct aspects of WM, and their associations with early clinical phenotypes.

The use of polygenic risk scores to identify phenotypes associated with genetic risk of bipolar disorder and depression: A systematic review

BACKGROUND

Identifying the phenotypic manifestations of increased genetic liability for depression (MDD) and bipolar disorder (BD) can enhance understanding of their aetiology. The polygenic risk score (PRS) derived using data from genome-wide-association-studies can be used to explore how genetic risk is manifest in different samples.

AIMS

In this systematic review, we review studies that examine associations between the MDD and BD polygenic risk scores and phenotypic outcomes.

METHODS

Following PRISMA guidelines, we searched EMBASE, Medline and PsycINFO (from August 2009 - 14th March 2016) and references of included studies. Study inclusion was based on predetermined criteria and data were extracted independently and in duplicate.

RESULTS

Twenty-five studies were included. Overall, both polygenic risk scores were associated with other psychiatric disorders (not the discovery sample disorder) such as depression, schizophrenia and bipolar disorder, greater symptom severity of depression, membership of a creative profession and greater educational attainment. Both depression and bipolar polygenic risk scores explained small amounts of variance in most phenotypes (< 2%).

LIMITATIONS

Many studies did not report standardised effect sizes. This prevented us from conducting a meta-analysis.

CONCLUSIONS

Polygenic risk scores for BD and MDD are associated with a range of phenotypes and outcomes. However, they only explain a small amount of the variation in these phenotypes. Larger discovery and adequately powered target samples are required to increase power of the PRS approach. This could elucidate how genetic risk for bipolar disorder and depression is manifest and contribute meaningfully to stratified medicine.

Gender differences in C-reactive protein and homocysteine modulation of cognitive performance and real-world functioning in bipolar disorder

BACKGROUND

Cognitive and psychosocial impairment has been associated with increased levels of C-reactive protein (CRP) and homocysteine in bipolar disorder, but gender differences have seldom been studied.

METHODS

Two hundred and twenty-four bipolar outpatients were included. Cognitive performance was assessed through the Screen for Cognitive Impairment in Psychiatry (SCIP). Psychosocial functioning was evaluated using the Functioning Assessment Short Test (FAST) and the General Assessment of Functioning (GAF). Homocysteine and CRP levels were determined. Separate analyses were performed by gender. Partial correlations were calculated to test for associations between biomarkers and cognitive and psychosocial functioning. Hierarchical multiple regression was used to assess factors predicting cognitive and psychosocial functioning. Covariates were: age, education, duration of illness, hospital admissions, depressive symptoms, tobacco consumption, and BMI.

RESULTS

A better performance was noted in women in delayed verbal learning (p = 0.010), along with better occupational functioning (p = 0.027) and greater leisure time impairment (p = 0.034). In men, CRP and homocysteine levels were associated with psychosocial dysfunction (interpersonal relationships and financial functioning, respectively). In women, CRP levels correlated with cognitive performance (SCIP total raw score, immediate and delayed verbal learning, and verbal fluency). CRP was a predictor of cognitive performance in women only.

LIMITATIONS

The choice of the cognitive scale and covariates and the lack of a control group may be the main limitations.

CONCLUSIONS

A gender difference was found in biomarker modulation of cognition and psychosocial functioning. A gender-based approach to cognition and real-world functioning should be considered in bipolar disorder to ensure an optimal outcome.

Staging in bipolar disorder: from theoretical framework to clinical utility

Illness staging is widely utilized in several medical disciplines to help predict course or prognosis, and optimize treatment. Staging models in psychiatry in general, and bipolar disorder in particular, depend on the premise that psychopathology moves along a predictable path: an at-risk or latency stage, a prodrome progressing to a first clinical threshold episode, and one or more recurrences with the potential to revert or progress to late or end-stage manifestations. The utility and validity of a staging model for bipolar disorder depend on its linking to clinical outcome, treatment response and neurobiological measures. These include progressive biochemical, neuroimaging and cognitive changes, and potentially stage-specific differences in response to pharmacological and psychosocial treatments. Mechanistically, staging models imply the presence of an active disease process that, if not remediated, can lead to neuroprogression, a more malignant disease course and functional deterioration. Biological elements thought to be operative in bipolar disorder include a genetic diathesis, physical and psychic trauma, epigenetic changes, altered neurogenesis and apoptosis, mitochondrial dysfunction, inflammation, and oxidative stress. Many available agents, such as lithium, have effects on these targets. Staging models also suggest the utility of stage-specific treatment approaches that may not only target symptom reduction, but also impede illness neuroprogression. These treatment approaches range from prevention for at-risk individuals, to early intervention strategies for prodromal and newly diagnosed individuals, complex combination therapy for rapidly recurrent illness, and palliative-type approaches for those at chronic, late stages of illness. There is hope that prompt initiation of potentially disease modifying therapies may preclude or attenuate the cognitive and structural changes seen in the later stages of bipolar disorder. The aims of this paper are to: a) explore the current level of evidence supporting the descriptive staging of the syndromal pattern of bipolar disorder; b) describe preliminary attempts at validation; c) make recommendations for the direction of further studies; and d) provide a distillation of the potential clinical implications of staging in bipolar disorder within a broader transdiagnostic framework.

Claiming the validity of scientific evidence in post-truth times

This letter is written in response to a review recently published in the journal. The aim is to highlight a potential methodological limitation common to many studies comparing bipolar patients with few previous episodes versus those with multiple episodes, and in which the results are interpreted as indicating the longitudinal course of the illness.

The effect of premorbid intelligence on neurocognitive and psychosocial functioning in bipolar disorder

BACKGROUND

The aim of this study was to assess if premorbid IQ moderates the association between measures of clinical severity and neurocognitive or psychosocial functioning in euthymic patients with bipolar disorder.

METHODS

One hundred and nineteen outpatients and forty healthy controls were included. The length of illness, number of previous hypo/manic and depressive episodes, episode density, and history of psychosis assessed clinical severity. Performances in verbal memory, attention, and executive functions, as well as level of psychosocial functioning were used as outcomes.

RESULTS

The negative relationship between number of hypo/manic episodes and performance in executive functions decreased as a function of higher values of premorbid IQ. No other influences of premorbid IQ were found in the association between clinical severity measures and neurocognitive and psychosocial functioning.

CONCLUSIONS

Premorbid IQ might moderate the relationship between the number of hypo/manic episodes and executive functioning in bipolar disorder. Possible interpretations of this finding are discussed.

Functional Outcome in the Middle Course of Bipolar Disorder: A Longitudinal Study

The aim of this study was to assess the long-term functional outcome of patients with bipolar disorder (BD). At baseline and after a follow-up period of at least 48 months, three measures of functioning were administered: psychosocial functioning (GAF), employment status (full-time, part-time, and unemployment/disability), and a self-reported measure of functional recovery. At baseline, patients with more than five previous affective episodes exhibited poorer outcomes on all measures of functioning than patients with less than five previous episodes. However, along a mean follow-up period of 77 months, measures of functioning tended to remain stable or improved slightly. These results highlight the limitation of studies comparing measures of functioning between patients with many and few episodes to evaluate functional outcome. Likewise, these preliminary results do not support the hypothesis that functional outcome deteriorates over the course of BD.

Neuropsychological profiles of major depressive disorder and bipolar disorder during euthymia. A systematic literature review of comparative studies

Bipolar disorder and major depressive disorder have been shown to be associated with neurocognitive abnormalities during periods of clinical remission. However, at present, there is no consensus on whether these disorders have distinctive cognitive profiles. The aim of this study was to provide an updated systematic review of studies comparing neuropsychological functioning between bipolar disorder and major depressive disorder during remission. Main findings included the following: 1) no differences regarding performances in measures of attention and processing speed, executive functions and theory of mind were found between both patient groups and 2) regarding verbal memory, preliminary evidence points towards a more defective performance in patients with bipolar disorder than those with major depressive disorder. However, several variables with negative impact on cognition (medication status, age at onset, premorbid IQ, bipolar subtype, among others) were not adequately controlled in most studies. In conclusion, evidence from studies exploring neuropsychological profiles in bipolar disorder and major depressive disorder could not provide clues to differentiate these mood disorders. Larger studies with adequate control of confounding variables would be necessary to elucidate if the finding of more defective verbal memory performance in bipolar disorder is truly explained by distinct underlying mechanisms.

Evidence for clinical progression of unipolar and bipolar disorders

OBJECTIVE

It is a widely held belief that affective disorders are progressive of nature; however, some recent reviews have questioned this belief. The objective of the present systematic literature review was to present evidence for associations between number of affective episodes and (i) the risk of recurrence of episodes, (ii) probability of recovery from episodes, (iii) severity of episodes, (iv) the threshold for developing episodes, and (v) progression of cognitive deficits in unipolar and bipolar disorders.

METHOD

A systematic review comprising an extensive literature search conducted in Medline, Embase, and PsychInfo up to September 2016 and including cross-references from identified papers and reviews.

RESULTS

Most of the five areas are superficially investigated and hampered by methodological challenges. Nevertheless, studies with the longest follow-up periods, using survival analysis methods, taking account of the individual heterogeneity all support a clinical progressive course of illness. Overall, increasing number of affective episodes seems to be associated with (i) increasing risk of recurrence, (ii) increasing duration of episodes, (iii) increasing symptomatic severity of episodes, (iv) decreasing threshold for developing episodes, and (v) increasing risk of developing dementia.

CONCLUSION

Although the course of illness is heterogeneous, there is evidence for clinical progression of unipolar and bipolar disorders.

A Longitudinal Mirror-Image Assessment of Morbidity in Bipolar Disorder

Background

Evidence about the clinical course of bipolar disorder is inconsistent and limited. The aim of this study was to assess changes in morbidity in patients with bipolar disorder along a mean follow-up period of 80 months.

Methods

Based on a mirror-image design, the follow-up period of each patient was divided into two halves. Then, three measures of morbidity — number of affective episodes, time spent ill, and cycle length — were recorded and compared between each half of the follow-up period.

Results

On average, there was a trend to a smaller amount of time spent with subclinical symptomatology during the second half of the follow-up period. In contrast, there were no differences in terms of number of episodes, time spent with clinical symptoms, or cycle length between the first and second half of the follow-up period. A subgroup analysis identified 21.9% of patients with consistent data of a worsening during follow-up.

Conclusions

The results suggest that, on average, there is stability or slight improvement of clinical morbidity over the course of BD. Then, worsening of the clinical course may be a feature of a subgroup of patients rather than an inherent characteristic of the disorder. These subgroups or patient profiles could represent an opportunity for further studies to assess clinical, pathophysiologic, and therapeutic features associated with them.

Stability of facial emotion recognition performance in bipolar disorder

The aim of this study was to assess the performance in emotional processing over time in a sample of euthymic patients with bipolar disorder (BD). Performance in the facial recognition of the six basic emotions (surprise, anger, sadness, happiness, disgust, and fear) did not change during a follow-up period of almost 7 years. These preliminary results suggest that performance in facial emotion recognition might be stable over time in BD.