How Does Dementia Begin to Manifest in Bipolar Disorder? A Description of Prodromal Clinical and Cognitive Changes

Abstract

Background:

Older adults with bipolar disorder (BD) have increased dementia risk, but signs of dementia are difficult to detect in the context of pre-existing deficits inherent to BD.

Objective:

To identify the emergence of indicators of early dementia in BD.

Methods:

One hundred and fifty-nine non-demented adults with BD from the National Alzheimer’s Coordinating Center (NACC) data repository underwent annual neuropsychological assessment up to 14 years (54.0 months average follow-up). Cognitive performance was examined longitudinally with linear mixed-effects models, and yearly differences between incident dementia cases and controls were examined in the six years prior to diagnosis.

Results:

Forty participants (25.2%) developed dementia over the follow-up period (‘incident dementia cases’). Alzheimer’s disease was the most common presumed etiology, though this was likely a result of sampling biases within NACC. Incident dementia cases showed declining trajectories in memory, language, and speeded attention two years prior to dementia onset.

Conclusion:

In a sample of BD patients enriched for Alzheimer’s type dementia, prodromal dementia in BD can be detected up to two years before onset using the same cognitive tests used in psychiatrically-healthy older adults (i.e., measures of verbal recall and fluency). Cognition in the natural course of BD is generally stable, and impairment or marked decline on measures of verbal episodic memory or semantic retrieval may indicate an early neurodegenerative process.

Keywords

Aging cognitive dysfunction neuropsychology psychiatry

INTRODUCTION

Cognitive deficits are a core feature of bipolar disorder (BD) [1]. Executive functions and episodic memory are most commonly affected, although global cognition is also typically impaired [2, 3]. Cognitive impairments are present even in the absence of mood symptoms, as well as in first-degree relatives [4], suggesting that they might be endophenotypes of the disease. They are consistent and persistent in the natural course of BD, although generally mild (up to 1 standard deviation [SD] below normal) [3 , 6]. In addition to deficits encompassed within the natural course of BD, considerable evidence from systematic reviews [7, 8] suggests that BD increases risk for dementia later in life, most frequently Alzheimer’s disease (AD) [1]. This does not appear to be due to worsening of pre-existing cognitive deficits [7]. It is possible that dementia risk is increased in this population due to chronic use of medications used to treat BD, many of which can adversely affect cognition [9]; however, the bulk of the available evidence does not provide strong support for this hypothesis. For example, a recent large population study of 66 million people indicates that dementia risk is independent of medication use [10], and meta-analysis of 70,000 bipolar patients found that only a very small proportion of neurodegenerative risk was accounted for by medication use [11]. A relatively more plausible explanation is that dementia risk may be related to BD’s potential to lower cognitive reserve through increasing allostatic load [12]. Regardless of the mechanism at play, dementia prevalence in BD is significantly higher (20–25%[13, 14]) than prevalence in the general population (7%in adults aged 60 or older [15]).

Despite this apparent heightened risk, studies aiming to increase knowledge about dementia syndromes systematically exclude subjects with psychiatric conditions [16]. In addition, the majority of published studies examining cognition in adult BD patients include only subjects younger than age 50 [17]. Thus, relatively little is known about the emergence of cognitive decline in BD later in life, or how clinicians should go about evaluating cognitive performance in the context of pre-existing deficits. Confusion between the ‘core’ deficits (i.e., those that are part of the natural history of BD) and those that signal a neurodegenerative process has been highlighted previously [18, 19]. Therefore, establishing features signaling the onset of neurodegenerative changes in BD would be helpful to inform the differential diagnosis at a critical period when early therapeutic intervention may be possible and important healthcare decisions must be made.

Previous studies have called for the identification of neuropsychological markers of prodromal dementia in individuals with BD [1, 2]. The few existing investigations of cognitive symptoms in older cohorts have focused on comparing elderly individuals with BD to age-matched controls without BD who are cognitively healthy (e.g., [20, 21]) or cognitively impaired (e.g., [22]). Only one previous study has directly compared cognitive performance in BD patients with and without dementia [2]; in this study, the authors assessed the use of the Cambridge Cognition Examination, Mini-Mental State Examination (MMSE), animal fluency, and clock drawing to distinguish BD with mild AD from BD alone, and reported good sensitivity and specificity for all tests. However, it remains unclear whether cognitive measures have value to identify BD subjects who are in the earliest stages of a neurocognitive disorder, given that all participants in the aforementioned study already had dementia. Furthermore, participants with non-AD dementias were excluded.

Here, we used longitudinal neuropsychological data collected over a 14-year span to document the emergence of early features of dementia in BD, relative to those that constitute the natural course of cognitive changes in individuals with BD who do not develop dementia. The approach was exploratory and data-driven, and therefore did not include any a priori hypotheses.

METHODS

Participants

This study used data from the National Alzheimer’s Coordinating Center (NACC; https://naccdata.org), a prospective cohort study consisting of participants recruited from several Alzheimer’s Disease Centers (ADC) across the United States, and includes individuals with dementia, mild cognitive impairment (MCI), or normal cognition. Data for this study were for visits conducted between August 2005 and the December 2020 data freeze, and included any participant with self-reported or clinician-diagnosed BD. Unfortunately, the NACC data repository does not include information on BD type (i.e., I, II, or unspecified).

Participants were 347 adults with BD seen at approximately yearly intervals. Individuals with no follow-up after their first visit (n = 126) or those who were demented on admission (n = 107) were excluded; 45 participants met both exclusion criteria. The remaining 159 participants were seen up to 169 months (5.2 total visits including baseline, or 54.0 months, on average). Seventy-eight cases (49%) had self-reported BD, and the remaining 81 (51%) were clinician-diagnosed. At baseline, 68 of the 159 participants were determined by the evaluating ADC clinician to have MCI, 19 were ‘impaired, no MCI’, and the remaining 72 were cognitively normal. Participants were classified as ‘incident dementia cases’ if they developed dementia at any point during the study. Dementia diagnoses were based on clinical judgement using DSM-IV criteria [23] (or similar/modified criteria; NACC acknowledges that diagnostic criteria may have varied between ADC sites). DSM-IV criteria include: 1) deficits in memory and at least one other cognitive domain; 2) significant impairment in social or occupational functioning due to cognitive impairments; 3) gradual onset and progressive course; 4) absence of contributing neurological or systemic conditions, delirium or other mental disorder. Forty participants (25.2%) developed dementia during the follow-up period; the remaining 119 cases (74.8%) remained free of dementia for the duration of the study (up to 14 years) and constituted the control group. At their last visit, 39 of these 119 control participants had MCI (most of these had also been classified as ‘MCI’ at baseline; one had been ‘impaired, not MCI’ and five had been cognitively normal), 17 were considered to be ‘impaired, not MCI’ (all but five had also been impaired at baseline). The remaining 63 were cognitively normal at their last visit (Fig. 1).

Fig. 1

Classification changes from first to last visit, with average length of follow-up in months. MCI, mild cognitive impairment; Mos, months.

All contributing ADCs obtained participant consent and received approval from their individual Institutional Review Board prior to submitting data to NACC.

Risk factors for cognitive decline

The presence of neuropsychiatric symptoms [24] and vascular risk factors [25] has been associated with cognitive decline in non-BD populations. To determine the impact of these factors on dementia risk in this sample, data were used from the Neuropsychiatric Inventory Questionnaire (NPI-Q) [26], which was completed by a knowledgeable informant. Vascular risk factors were assessed using the Hachinski scale [27]. Information about tobacco use (total years smoked) and alcohol abuse (absent, recent/active, remote/inactive) was collected. Mood-stabilizing medications, such as lithium, are also known to impact cognition [28] and can moderate dementia risk [29], and thus were also considered as relevant factors to consider. Medication use within two weeks of assessment was documented in the present sample; unfortunately, lifetime exposure was not. All mood-stabilizing medications, as well as antipsychotics, benzodiazepines, and other antiepileptic drugs that can be used off-label as mood stabilizers, were considered.

Cognitive outcomes

All participants were administered the ADC Uniform Data Set (UDS) neuropsychological battery [30], which included the Wechsler Memory Scale Revised (WMS-R) forward and backward digit spans, WMS-R digit-symbol substitution (DSS), trails A and B, WMS-R Logical Memory Story A, semantic fluency (animals and vegetables) and confrontation naming. The MMSE was used as a brief measure of general cognition. Variables coded as ‘missing due to a cognitive/behavior problem’ were replaced with the lowest allowable score for that test. All other missing variables were left missing. Scores were then standardized to Z-scores using age-, sex-, and education-adjusted normative data derived from 3,268 cognitively normal participants from NACC published previously [31], where Z = 0 provides a benchmark for expected performance in healthy older adults (with an associated standard deviation of 1), thus precluding the need for a non-psychiatric normal control group. In addition to formal cognitive testing, participants were asked about subjective reports of memory decline relative to previous abilities (one subject had missing data for this question). They were also queried about the predominant cognitive symptom that was first recognized as a change from previous levels.

Structural neuroimaging

The NACC data repository included available structural magnetic resonance imaging (MRI) data on a small subset of 12 BD participants, including 5 incident dementia cases (4 males, 1 female) and 7 controls (4 males, 3 females). These images were analyzed to obtain supplementary insights into the possible neurological contributions to cognitive decline in BD. To this end, intracranial volumes were acquired using automatic segmentation in SPM12 (http://www.fil.ion.ucl.ac.uk/spm) which calculated white matter, grey matter, and cerebrospinal fluid volumes for each participant. These volumes were summed and corrected for voxel size to ensure consistency across participants. In addition, because AD is the most frequent type of dementia seen in BD [1] and is characterized by significant hippocampal atrophy, bilateral, manual segmentations of the hippocampi were traced for each participant using ITK-SNAP [32] (http://www.itksnap.org). A well-established protocol for manual segmentation of the hippocampus in T1-weighted images was used to train both the primary and secondary rater [33]. Tracing reliability was assessed by having a second rater trace four of the 12 brains (two from each group). Volumes from the four tracings were highly correlated between raters (r (2) = 0.926, p = 0.247) suggesting a high degree of inter-rater reliability. Left and right hippocampal volumes were summed and corrected for voxel size to produce total hippocampal formation volumes for each participant.

Statistical analyses

All analyses were two-tailed, with alpha set at 0.05 except where stated otherwise. The incident dementia cases and dementia-free controls were first compared in terms of age and years of education using Mann-Whitney U, because the data were non-normally distributed (Shapiro-Wilk p < 0.001 for both variables). Sex and subjective impression of cognitive change at any time during the study (yes/no) were compared between groups using chi-square (χ²).

Neuropsychiatric symptoms, vascular risk factors, tobacco, alcohol and medication use were compared longitudinally using linear mixed effects models with Group (incident dementia cases, controls) as a fixed effect. Because all medications (n = 29) were considered independently in separate statistical analyses, the significance threshold was set at p < 0.002 using Bonferroni correction for multiple comparisons (0.05/29). A Group×Time interaction term was added to the model with neuropsychiatric symptoms in order to determine whether worsening neuropsychiatric symptoms were associated with incident dementia.

Cognitive change was examined as a function of the time interval (in months) between each assessment and ‘Time0’, ranging from 0 to 169. ‘Time0’ was always the date of dementia diagnosis for incident dementia cases, and last follow-up visit for controls. Of note, the earliest date of dementia diagnosis was always used, even if diagnosis subsequently reverted to MCI (n = 8; four of these cases were ultimately re-diagnosed with dementia at a later visit). Linear mixed effects models were used to establish the relationship between longitudinal performance on each cognitive test and dementia status. This statistical approach allows for correlated repeated measures within subjects, while permitting the inclusion of participants with missing data at certain time points [34]. Cognitive variables were entered individually into separate models with Time and Group as fixed effects, and covarying for sex as a fixed effect use due to a priori group differences (see Results, below). Group×Time interaction terms were added to all models and were the main outcome of interest. Two participants (both incident cases) were dropped from these analyses because they lacked cognitive data at all time points, which were coded as ‘Missing due to: Other problem’ (i.e., not due to a physical, cognitive or behavioral problem, nor due to verbal refusal).

Hippocampal atrophy was examined using a one-way analysis of covariance (ANCOVA) accounting for age, intracranial volume, and the number of days between Time0 and the date of MRI acquisition. Additionally, a group effect was discovered in the number of days between Time0 and the date of MRI acquisition. Consequently, group means were removed from this variable by subtracting the group mean from the amount of days to Time0 for each participant. This was done to remove any inter-group variance while still maintaining a marginal amount of intra-group variance. Sex was not included as a covariate in these analyses because there were no sex differences in the neuroimaging sample (χ² = 0.686, p = 0.408).

Results

Participant characteristics

Participants’ sociodemographic characteristics are summarized in Table 1. The 40 incident cases developed dementia after a mean of 42.3 months (SD = 34.0), and the 119 controls remained free of dementia for a mean follow-up duration of 57.9 months (SD = 38.1); follow-up length was significantly longer for controls (Z = –2.616, p = 0.009). Participants in the incident dementia group were roughly 3.6 years older at their first visit than control participants (Z = –2.308; p = 0.021), but age at Time0 was statistically similar between the incident dementia group (73.6±12.3 years) and the control group (71.5±11.2 years; Z = –1.599, p = 0.110). Education was also comparable between the incident dementia group (15.3±3.1 years) and the control group (16.5±2.7 years; Z = –1.905, p = 0.057). The incident dementia group included a significantly greater proportion of males (65%) than the control group (37%; χ²= 9.541, p = 0.002). All but one incident dementia case reported concern about their memory at some point during follow-up (97.5%) compared to 71.4%of controls (χ²= 11.853, p = 0.001).

Table 1

Mean (SD) characteristics of incident dementia cases and controls

	Incident dementia cases (n = 40)	Controls (n = 119)	Statistic	p
Age at first visit	70.10 (11.61)	66.54 (10.78)	Z = –2.308	0.021
Age at Time 0	73.60 (12.24)	71.45 (11.21)	Z = –1.599	0.110
Years of education	15.33 (3.12)	16.51 (2.73)	Z = –1.905	0.057
Sex (%males)	65.0%	37.0%	χ²= 9.541	0.002
Mean follow-up length (months)	42.33 (33.95)	57.93 (38.10)	Z = –2.616	0.009
Subjective sense of decline at any time (%yes)	97.5%	71.4%	χ²= 11.853	0.001
MMSE at first visit^a	27.38 (2.30)	28.69 (1.65)	F = 58.166^b	<0.001
NPI-Q symptom severity at first visit^a	78.50 (14.33)	83.21 (13.19)	F = 11.938^b	0.001
NPI-Q symptom distress at first visit^a	2.76 (2.24)	1.95 (2.05)	F = 13.072^b	<0.001
Hachinski score at first visit^a	0.94 (1.39)	0.76 (1.03)	F = 1.066^b	0.303
Smoking history (years, reported at first visit^a)	10.75 (15.60)	11.88 (15.68)	F = 0.532^b	0.467
Alcohol abuse (%never, reported at first visit^a)	76.3%	79.0%	F = 0.018^b	0.893

SD, standard deviation; NPI-Q, Neuropsychiatric Inventory Questionnaire. ^aFor parsimony, means refer to only data collected at first visit, but linear mixed effects models included data collected at all time points. ^bStatistic refers to the F value associated with the main effect of Group in longitudinal linear mixed models.

Risk factors for cognitive decline

Throughout the follow-up period, the presence of neuropsychiatric symptom distress (Estimate_Group = –1.143, p < 0.001) and severity (Estimate_Group = 6.957, p = 0.001) was statistically different between groups, but the Group×Time interactions terms in the NPI-Q distress model (Estimate_G_×T = –0.004, p = 0.376) and the NPI-Q severity model (Estimate_G_×T = 0.021, p = 0.473) were not. Vascular risk factors were similar between incident cases and controls (Estimate_Group = 0.269, p = 0.303), and there was no Group×Time interaction (Estimate_G_×T = 0.004, p = 0.276). The groups did not differ on smoking (Estimate_Group = 2.302, p = 0.467; Estimate_G_×T = 0.005, p = 0.489) or alcohol abuse (Estimate_Group = –0.020, p = 0.893; Estimate_G_×T = 0.001, p = 0.195) nor on two-week prior use of any psychotropic medication after correcting for multiple comparisons (all main effects and interactions p > 0.002).

Cognitive outcomes

Of the 40 incident dementia cases, 28 (70.0%) reported memory as the predominant symptom first recognized as a decline in cognition, eight (20.0%) reported executive difficulties (judgment, planning, problem-solving), and four (10.0%) reported language symptoms. BD was determined to be the primary etiology contributing to cognitive impairment in one participant (0.3%), who also had AD as a contributing factor. Twenty-two participants (55.0%) had AD as a primary etiologic diagnosis, four (10.0%) had frontotemporal lobar degeneration, and four (10.0%) had Lewy body dementia. One participant had alcohol-related dementia, one had vascular dementia, one had normal-pressure hydrocephalus, and one had medication-induced cognitive impairment. Two participants had non-BD psychiatric factors listed as the primary etiologic factors for cognitive impairment, with cerebrovascular disease as an additional contributing factor in one. In one participant, etiology was undetermined. These etiologic diagnoses were presumptive and determined by clinicians within individual ADCs based on clinical presentation.

Sex-adjusted linear mixed model results are illustrated in Fig. 2, and all estimates are reported in Table 2. A main effect of Group was apparent in all models, and a main effect of Time was apparent in all models except backward digit span or DSS. A significant Group×Time interaction was present in models involving immediate and delayed story recall, forward digit span, animal fluency, vegetable fluency, trails A and B, and confrontation naming. No interactions were significant in backwards digit span or DSS.

Fig. 2

Cognitive change nine years preceding a diagnosis of dementia (incident dementia cases, dotted line) or last follow-up (controls, solid line). All models adjusted for Sex. ^*Statistically significant group difference after adjusting for multiple comparisons (p < 0.005).

Table 2

Linear mixed effects model results

	Estimate^a	Std. Error	t statistic	p	95%CI
Short story immediate recall
Group	1.536	0.253	6.071	< 0.001	1.036—2.036
Months to Time0	–0.009	0.003	–3.003	0.003	–0.015 — –0.003
Group×Time	0.010	0.003	3.035	0.003	0.004—0.017
Short story delayed recall
Group	1.375	0.242	5.673	< 0.001	0.896—1.854
Months to Time0	–0.007	0.003	–2.516	0.012	–0.013 — –0.002
Group×Time	0.008	0.003	2.431	0.015	0.001—0.014
Forward digit span
Group	0.724	0.237	3.055	0.003	0.255—1.192
Months to Time0	–0.007	0.002	–2.687	0.007	–0.011 — –0.002
Group×Time	0.006	0.003	2.346	0.019	0.001—0.012
Backward digit span
Group	0.781	0.238	3.279	0.001	0.310—1.252
Months to Time0	–0.005	0.003	–1.963	0.050	–0.011—0.000
Group×Time	0.006	0.003	1.857	0.064	0.000—0.012
Animal fluency
Group	0.716	0.178	4.023	< 0.001	0.365—1.068
Months to Time0	–0.009	0.002	–4.318	0.000	–0.013 — –0.005
Group×Time	0.006	0.002	2.759	0.006	0.002—0.011
Vegetable fluency
Group	0.625	0.183	3.418	0.001	0.264—0.986
Months to Time0	–0.011	0.002	–5.034	0.000	–0.016 — –0.007
Group×Time	0.006	0.002	2.402	0.017	0.001—0.011
Trails A
Group	1.018	0.243	4.189	< 0.001	0.538—1.497
Months to Time0	–0.008	0.003	–2.914	0.004	–0.014 — –0.003
Group×Time	0.009	0.003	2.731	0.006	0.002—0.015
Trails B
Group	1.108	0.251	4.424	< 0.001	0.614—1.603
Months to Time0	–0.013	0.003	–4.029	< 0.001	–0.002 — –0.007
Group×Time	0.009	0.004	2.435	0.015	0.002—0.016
Digit symbol substitution
Group	1.089	0.235	4.622	< 0.001	0.623—1.554
Months to Time0	–0.002	0.003	–0.691	0.490	–0.007—0.003
Group×Time	0.002	0.003	0.826	0.409	–0.003—0.008
Confrontation naming
Group	1.597	0.298	5.355	< 0.001	1.007—2.186
Months to Time0	–0.010	0.003	–3.083	0.002	–0.017 — –0.004
Group×Time	0.013	0.004	3.485	0.001	0.006—0.020

All models adjusted for sex. ^aEstimates can be interpreted as follows: The unique effect of Time (‘Months to Time0’) on cognitive performance (always in Z score units, i.e., SD) for either group can be interpreted as Estimate_Time + (Estimate_Time_×Group×Group), where Group = 0 are controls and Group = 1 are incident dementia cases. For example, control participants’ confrontation naming scores change at a rate of –0.010 SD (i.e., –0.010 –0.013^*0) per month, whereas those in the incident dementia group decline at a rate of –0.023 SD per month (i.e., –0.010 –0.013^*1).

To determine the earliest point at which between-group differences appeared in models yielding a significant interaction term, yearly differences between incident cases and controls were examined using sex-adjusted ANCOVA. These analyses included only data collected between Time-9 and Time0 because of insufficient data at prior time points. Effect sizes are reported as partial eta squared (η_p), where values 0.01, 0.06, and 0.14 are considered small, medium and large, respectively [35]. There was a single group difference in trails B at seven years before diagnosis (F = 4.872, p = 0.008), where the average Z scores for incident dementia cases and controls were M = –2.01 (SD = 1.37) and M = –0.18 (SD = 1.11), respectively (p = 0.006, η_p = 0.250). Models in subsequent years, however, were no longer significant (all corrected models p > 0.05). Two years prior to dementia diagnosis, corrected models were significant for immediate (F = 7.295, p = 0.001) and delayed story recall (F = 8.298, p < 0.001), vegetable fluency (F = 13.490, p < 0.001) and trails A (F = 4.830, p = 0.007). On average, the incident dementia cases obtained Z scores on these tests that fell respectively 1.68 (p = 0.001, η_p = 0.304), 1.72 (p = 0.001, η_p = 0.287), 0.79 (p = 0.007, η_p = 0.198) and 0.97 standard deviations (p = 0.002, η_p = 0.266) below those of controls. One year prior to dementia onset, all these same tests showed significant between-group differences, with the new addition of animal fluency (F = 2.950, p = 0.045). At this time point, the incident dementia cases’ average Z scores (M = –1.07, SD = 1.06) fell significantly below those of controls (M = –0.43, SD = 1.07; p = 0.008, η_p = 0.175). At Time0, the corrected model for confrontation naming became significant (F = 3.006, p = 0.041), with significantly lower mean Z scores in the incident dementia group (M = –1.69, SD = 2.40) than in controls (M = –0.56, SD = 1.42; p = 0.011, η_p = 0.146).

Neuroimaging

Intracranial volumes were compared across groups to ensure consistency of the automatic tracings. Intracranial volumes did not differ significantly between the incident dementia cases (M = 1,540 cm³, SD = 261 cm³) and controls (M = 1,476 cm³, SD = 211 cm³; t (10) = 0.477, p = 0.644). With the limited amount of imaging data provided for participants, hippocampal atrophy comparisons were largely underpowered. Therefore, hippocampal atrophy was only compared at the level of marginal means, providing a purely descriptive approach. A 7.4%reduction in hippocampal volume was found in the incident dementia cases (M = 3.255 cm³, SE = 0.274 cm³) compared to controls (M = 3.516 cm³, SE = 0.230 cm³) while controlling for age, intracranial volume, and number of days between Time0 and the date of MRI acquisition. This is on par with previous literature suggesting that mild to moderate dementia is associated with a 5%to 10%reduction in hippocampal volume, respectively [36].

DISCUSSION

To distinguish cognitive deficits inherent to BD from those signaling incident dementia, this study examined a cohort of older adults with BD drawn from the NACC data repository. The NACC is an AD-sensitive database and was not designed to examine BD-specific outcomes. As a result, we interpret our results cautiously and acknowledge that these findings may not generalize to the broader BD population.

Over the course of the study, a quarter (25.2%) of the eligible sample eventually received a diagnosis of dementia, within the range of previously-reported dementia rates of 20–25%in BD [13, 14] and higher than the 7%population prevalence in non-psychiatric samples [15]. This group included predominantly males (65.0%), and almost all endorsed subjective concerns at some point during the study. Memory problems were the predominant symptoms first recognized as a change in cognition, and AD was the most common presumed etiology. This was tentatively supported by relatively smaller hippocampal volumes in a small subset of the incident dementia cases, although the very few subjects with available imaging data resulted in these analyses being severely underpowered. These data were included to provide some corroboration of underlying pathology, in line with recent research recommendations from the National Institutes of Aging and Alzheimer’s Association [37], but future work should examine MRI markers of neurodegeneration in BD more comprehensively. The high prevalence of AD in this sample is likely due to the fact that participants in the NACC data repository were recruited through ADCs and are biased towards having AD. The NACC emphasizes that its data cannot be used to estimate prevalence or incidence of dementia in external samples due to this bias.

While acknowledging the constraints of the sample, our results suggest that a standard neuropsychological assessment can be used to detect signs of dementia several years prior to dementia diagnosis in individuals with BD, and that neurodegenerative processes—primarily AD—in this group of patients resemble that of the general population (i.e., early primary disturbances in memory and language, with a subjective sense of decline). Though this finding may appear self-evident, to our knowledge, it has not been documented empirically by previous studies. We found that performance on all cognitive measures was associated with incident dementia, with short story recall and semantic fluency showing the earliest reliable between-group differences, up to two years before diagnosis. This is consistent with findings of other studies that have documented the temporal emergence of clinical dementia features in non-psychiatric samples, and have found that impairments in verbal episodic memory and semantic fluency are among the first detectable signs preceding dementia onset in MCI due to AD [38, 39]. In these prior studies, however, meaningful cognitive markers of future dementia typically appear earlier, around eight years prior to dementia onset [38] and sometimes as early as 12 years prior [39]. It is possible that very early clinical manifestations of dementia in BD are somewhat concealed by its psychiatric manifestations, making early detection difficult. More plausibly, our relatively small sample (particularly at early time points) was likely underpowered to detect group differences at earlier assessments.

Importantly, performance on most neuropsychological tests was not impaired in BD controls who remained dementia-free throughout the study: at all time points, these patients’ performance hovered around Z = 0 (i.e., clinically normal) for most measures, particularly those involving memory and language (Fig. 1). Although this contrasts with reports of cognitive impairments being a core feature of BD [1], it is in line with other evidence suggesting that cognitive impairments within the natural course of BD are mostly mild [3 , 6]. It also may be a reflection of selection biases within the NACC sample: cognitively healthy volunteers in research studies at ADCs are highly educated, high-functioning individuals who are likely not representative of the broader BD population [40]. It will be imperative to replicate these findings, and to re-test the hypothesis that impairments on measures of episodic and semantic retrieval are predictors of accelerated longitudinal decline, in other BD samples. Currently, our results indicate that, in individuals with BD, scores falling below –1.0 SD on these measures are suggestive of subsequent cognitive worsening and dementia onset in the following 2 years. These patterns may be used to inform the differential diagnosis (i.e., natural course of cognition in BD versus neurodegenerative disease course) at a critical period when early therapeutic interventions can be initiated and important healthcare decisions must be made.

Factors such as diminished cognitive reserve, vascular risk factors, and exposure to pharmacological treatments have been proposed to account for increased dementia risk in BD [7, 12]. Although the present study was not designed to specifically test these associations, its results do not appear to support this view. No differences were found between the incident dementia cases and controls cases in terms of education [considered a cognitive reserve proxy, 47], or Hachinski score [a measure of vascular risk, 27]. Psychotropic drug use at the time of testing was also not different between groups, although due to limitations of the NACC dataset we were unable to examine whether the incident dementia cases differed from controls in their lifetime exposure to medications. Furthermore, statistical models revealed no interaction between group and time in terms of NPI-Q symptom severity or distress measures, indicating that dementia onset was not associated with worsening neuropsychiatric symptoms over time. Again, it is important to bear in mind the nature of the NACC dataset, and we recognize that individuals with severely worsening neuropsychiatric symptoms over time may have been lost to follow-up at their ADC and would not have been captured by these data. Eight incident dementia cases (20.0%) reverted from ‘dementia’ to ‘MCI’ status at least once during the study; this is in line with previous studies showing relatively less stability of dementia diagnoses in individuals with premorbid cognitive challenges [48].

Limitations

Data for this study were drawn from the NACC data repository, which inherently limits the generalizability of findings. AD is likely overrepresented in these data, and incident dementia cases within NACC may decline more significantly or at faster rates that incident dementia cases within the community. NACC included more than 40,000 participants enrolled as of the December 2020 data freeze, however only a small percentage of these had a diagnosis of BD, and ever fewer eventually developed dementia. As such, the relatively small sample of individuals retained for analysis may be viewed as a limitation of this study. Average follow-up length was also relatively short in some cases, which may have prevented us from detecting meaningful cognitive change and erroneously classifying some individuals as ‘controls’.

Another important limitation concerns the reliability and validity of self-reported BD diagnoses. A considerable number of participants were cognitively impaired, which may have led to inaccurate reports of medical and psychiatric histories. Results should be replicated in more carefully characterized samples. Furthermore, a number of potentially important variables were also not collected in NACC, including lifetime exposure to different pharmacological treatments, number of previous affective episodes and hospitalizations, all of which are known to affect cognition in BD [3, 49]. Age of BD onset is also not available in the NACC dataset. This information would be helpful to confirm whether early- versus late-onset BD differ in their rates of cognitive impairment and dementia risk, as has been suggested previously [13].

Lastly, only a very small subsample of participants had available neuroimaging data, and these were obtained from a large collection of pooled scans from various centers with varying image acquisition protocols. Although the NACC protocol includes neuroimaging and cerebrospinal fluid (CSF) biomarker data collection to support the etiological diagnosis of dementia, these data were unfortunately not available for participants used in this study. Thus, it must be acknowledged that that the etiological diagnosis of dementia was presumptive. Future work should characterize the structural and functional neurological changes that are associated with incident dementia in BD, as well as the presence and relevance of CSF biomarkers. Results from one recent study suggest that cognitively impaired BD patients do not display AD-specific CSF biomarkers [50], however additional work is necessary to corroborate this finding and its association with cognition longitudinally.

CONCLUSIONS

Although epidemiologic evidence suggests that adults with BD may have increased dementia risk, to date, little is known about what neuropsychological changes may be harbingers of early neurodegeneration in these patients due to cognitive deficits being a core feature of BD independent of dementia. Our study provides preliminary evidence that some specific cognitive deficits can be used as predictive measures for early diagnosis of dementia in patients with BD. These findings are of significant relevance given that the planet’s population will include growing numbers of seniors, many of whom will have BD and may be in the early stages of dementia and need appropriate clinical management. Indeed, it has been estimated that one in four individuals with BD is over age 60 [51], and thus at high risk for neurodegenerative disease [13]. Future studies may rely on the findings provided in this study to further explore the issue of dementia in BD.

Footnotes

ACKNOWLEDGMENTS

The study subjects’ participation is gratefully acknowledged, as is database assistance from NACC coordinators and statistical assistance from Dr. Audrey Schnell.

Dr. Callahan’s work is funded through a Canada Research Chair in Adult Clinical Neuropsychology. The NACC database is funded by National Institute on Aging/National Institutes of Health (NIA/NIH) Grant U01 AG016976.

NACC data are contributed by the NIA funded Alzheimer’s Disease Centers (ADC): P30 AG019610 (PI Eric Reiman, MD), P30 AG013846 (PI Neil Kowall, MD), P50 AG008702 (PI Scott Small, MD), P50 AG025688 (PI Allan Levey, MD, PhD), P50 AG047266 (PI Todd Golde, MD, PhD), P30 AG010133 (PI Andrew Saykin, PsyD), P50 AG005146 (PI Marilyn Albert, PhD), P50 AG005134 (PI Bradley Hyman, MD, PhD), P50 AG016574 (PI Ronald Petersen, MD, PhD), P50 AG005138 (PI Mary Sano, PhD), P30 AG008051 (PI Steven Ferris, PhD), P30 AG013854 (PI M. Marsel Mesulam, MD), P30 AG008017 (PI Jeffrey Kaye, MD), P30 AG010161 (PI David Bennett, MD), P50 AG047366 (PI Victor Henderson, MD, MS), P30 AG010129 (PI Charles DeCarli, MD), P50 AG016573 (PI Frank LaFerla, PhD), P50 AG016570 (PI Marie-Francoise Chesselet, MD, PhD), P50 AG005131 (PI Douglas Galasko, MD), P50 AG023501 (PI Bruce Miller, MD), P30 AG035982 (PI Russell Swerdlow, MD), P30 AG028383 (PI Linda Van Eldik, PhD), P30 AG010124 (PI John Trojanowski, MD, PhD), P50 AG005133 (PI Oscar Lopez, MD), P50 AG005142 (PI Helena Chui, MD), P30 AG012300 (PI Roger Rosenberg, MD), P50 AG005136 (PI Thomas Montine, MD, PhD), P50 AG033514 (PI Sanjay Asthana, MD, FRCP), P50 AG005681 (PI John Morris, MD), and P50 AG047270 (PI Stephen Strittmatter, MD, PhD).

In addition, this work was supported by NIA grants awarded to D.A. Bennett #P30AG10161 and #R01AG17917.

Authors’ disclosures available online ().

References

Aprahamian

, Nunes

, Forlenza

(2013) Cognitive impairment and dementia in late-life bipolar disorder. Curr Opin Psychiatry 26, 120–123.

Aprahamian

, Radanovic

, Nunes

, Ladeira

, Forlenza

(2014) The use of the Clock Drawing Test in bipolar disorder with or without dementia of Alzheimer’s type. Arq Neuropsiquiatr 72, 913–918.

Forlenza

, Aprahamian

(2013) Cognitive impairment and dementia in bipolar disorder. Front Biosci (Elite Ed) 5, 258–265.

Arts

, Jabben

, Krabbendam

, van Os

(2008) Meta-analyses of cognitive functioning in euthymic bipolar patients and their first-degree relatives. Psychol Med 38, 771–785.

Robinson

, Nicol Ferrier

(2006) Evolution of cognitive impairment in bipolar disorder: A systematic review of cross-sectional evidence. Bipolar Disord 8, 103–116.

Thompson

, Gallagher

, Hughes

, Watson

, Gray

, Ferrier

, Young

(2005) Neurocognitive impairment in euthymic patients with bipolar affective disorder. Br J Psychiatry 186, 32–40.

Strejilevich

, Samamé

, Martino

(2015) The trajectory of neuropsychological dysfunctions in bipolar disorders: A critical examination of a hypothesis. J Affect Disord 175, 396–402.

Diniz

, Teixeira

, Cao

, Gildengers

, Soares

, Butters

, Reynolds III

(2017) History of bipolar disorder and the risk of dementia: A systematic review and meta-analysis. Am J Geriatr Psychiatry 25, 357–362.

Balanzá-Martínez

, Selva

, Martínez-Arán

, Prickaerts

, Salazar

, González-Pinto

, Vieta

, Tabarés-Seisdedos

(2010) Neurocognition in bipolar disorders–a closer look at comorbidities and medications. Eur J Pharmacol 626, 87–96.

10.

Harrison

, Luciano

(2020) Incidence of Parkinson’s disease, dementia, cerebrovascular disease and stroke in bipolar disorder compared to other psychiatric disorders: An electronic health records network study of 66 million people. Bipolar Disord, doi: 10.1111/bdi.13022.

11.

Faustino

, Duarte

, Chendo

, Castro Caldas

, Reimão

, Fernandes

, Vale

, Tinazzi

, Bhatia

, Ferreira

(2020) Risk of developing Parkinson disease in bipolar disorder: A systematic review and meta-analysis. JAMA Neurol 77, 192–198.

12.

Kapczinski

, Mwangi

, Cassidy

, Librenza-Garcia

, Bermudez

, Kauer-Sant’anna

, Kapczinski

, Passos

(2017) Neuroprogression and illness trajectories in bipolar disorder. Expert Rev Neurother 17, 277–285.

13.

Almeida

, McCaul

, Hankey

, Yeap

, Golledge

, Flicker

(2016) Risk of dementia and death in community-dwelling older men with bipolar disorder. Br J Psychiatry 209, 121–6.

14.

Nunes

, Forlenza

, Gattaz

(2007) Lithium and risk for Alzheimer’s disease in elderly patients with bipolar disorder. Br J Psychiatry 190, 359–360.

15.

Prince

, Bryce

, Albanese

, Wimo

, Ribeiro

, Ferri

(2013) The global prevalence of dementia: A systematic review and metaanalysis. Alzheimers Dement 9, 63–75.e2.

16.

Trivedi

, Humphreys

(2015) Participant exclusion criteria in treatment research on neurological disorders: Are unrepresentative study samples problematic? Contemp Clin Trials 44, 20–25.

17.

Cardoso

, Bauer

, Meyer

, Kapczinski

, Soares

(2015) Neuroprogression and cognitive functioning in bipolar disorder: A systematic review. Curr Psychiatry Rep 17, 75.

18.

Banga

, Gyurmey

, Matuskey

, Connor

, Kaplan

, Steffens

(2013) Late-life onset bipolar disorder presenting as a case of pseudo-dementia: A case discussion and review of literature. Yale J Biol Med 86, 235–244.

19.

Woudstra

, van de Poel-Mustafayeva

, van der Ploeg

M V

, de Vries

, van der Lek

RFR

, Izaks

(2014) Symptoms mimicking dementia in a 60-year-old woman with bipolar disorder: A case report. BMC Res Notes 7, 381.

20.

Gildengers

, Butters

, Seligman

, McShea

, Miller

, Mulsant

, Kupfer

, Reynolds

(2004) Cognitive functioning in late-life bipolar disorder. Am J Psychiatry 161, 736–738.

21.

Gildengers

, Mulsant

, Begley

, Mazumdar

, Hyams

, Reynolds III

, Kupfer

, Butters

(2009) The longitudinal course of cognition in older adults with bipolar disorder. Bipolar Disord 11, 744–752.

22.

Aprahamian

, Ladeira

, Diniz

, Forlenza

, Nunes

(2014) Cognitive impairment in euthymic older adults with bipolar disorder: A controlled study using cognitive screening tests. Am J Geriatr Psychiatry 22, 389–397.

23.

American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, American Psychiatric Publishing, Washington, DC.

24.

Geda

, Roberts

, Mielke

, Knopman

, Christianson

TJH

, Pankratz

, Boeve

, Sochor

, Tangalos

, Petersen

, Rocca

(2014) Baseline neuropsychiatric symptoms and the risk of incident mild cognitive impairment: A population-based study. Am J Psychiatry 171, 572–581.

25.

Gorelick

, Scuteri

, Black

, DeCarli

, Greenberg

, Iadecola

, Launer

, Laurent

, Lopez

, Nyenhuis

, Petersen

, Schneider

, Tzourio

, Arnett

, Bennett

, Chui

, Higashida

, Lindquist

, Nilsson

, Roman

, Sellke

, Seshadri

, American Heart Association Stroke Council, Council on Epidemiology and Prevention, Council on Cardiovascular Nursing, Council on Cardiovascular Radiology and Intervention, and Council on Cardiovascular Surgery and Anesthesia (2011) Vascular contributions to cognitive impairment and dementia: A statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 42, 2672–2713.

26.

Cummings

, Mega

, Gray

, Rosenberg-Thompson

, Carusi

, Gornbein

(1994) The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology 44, 2308–2314.

27.

Hachinski

, Iliff

, Zilhka

, DuBoulay

, McAllister

, Marshall

, Russell

, Symon

(1975) Cerebral blood flow in dementia. Arch Neurol 32, 632–637.

28.

Wingo

, Wingo

, Harvey

, Baldessarini

(2009) Effects of Lithium on Cognitive Performance. J Clin Psychiatry 70, 1588–1597.

29.

Velosa

, Delgado

, Finger

, Berk

, Kapczinski

, de Azevedo Cardoso

(2020) Risk of dementia in bipolar disorder and the interplay of lithium: A systematic review and meta-analyses. Acta Psychiatr Scand 141, 510–521.

30.

Weintraub

, Salmon

, Mercaldo

, Ferris

, Graff-Radford

, Chui

, Cummings

, DeCarli

, Foster

, Galasko

, Peskind

, Dietrich

, Beekly

, Kukull

, Morris

(2009) The Alzheimer’s Disease Centers’ Uniform Data Set (UDS): The neuropsychologic test battery. Alzheimer Dis Assoc Disord 23, 91–101.

31.

Shirk

, Mitchell

, Shaughnessy

, Sherman

, Locascio

, Weintraub

, Atri

(2011) A web-based normative calculator for the uniform data set (UDS) neuropsychological test battery. Alzheimers Res Ther 3, 32.

32.

Yushkevich

, Piven

, Hazlett

, Smith

, Ho

, Gee

, Gerig

(2006) User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability. Neuroimage 31, 1116–1128.

33.

Frisoni

, Jack

CRJ

, Bocchetta

, Bauer

, Frederiksen

, Liu

, Preboske

, Swihart

, Blair

, Cavedo

, Grothe

, Lanfredi

, Martinez

, Nishikawa

, Portegies

, Stoub

, Ward

, Apostolova

, Ganzola

, Wolf

, Barkhof

, Bartzokis

, DeCarli

, Csernansky

, deToledo-Morrell

, Geerlings

, Kaye

, Killiany

, Lehéricy

, Matsuda

, O’Brien

, Silbert

, Scheltens

, Soininen

, Teipel

, Waldemar

, Fellgiebel

, Barnes

, Firbank

, Gerritsen

, Henneman

, Malykhin

, Pruessner

, Wang

, Watson

, Wolf

, deLeon

, Pantel

, Ferrari

, Bosco

, Pasqualetti

, Duchesne

, Duvernoy

, Boccardi

(2015) The EADC-ADNI Harmonized Protocol for manual hippocampal segmentation on magnetic resonance: Evidence of validity. Alzheimers Dement 11, 111–125.

34.

Seltman

(2015) Mixed models. In Experimental Design and Analysis, Seltman HJ, ed., pp. 357–378.

35.

Cohen

(1988) Statistical power analysis for the behavioral sciences. In Lawrence Erlbaum Associates, Hillsdale, New Jersey, pp. 273–406.

36.

Vijayakumar

, Vijayakumar

(2013) Comparison of hippocampal volume in dementia subtypes. ISRN Radiol 2013, 174524.

37.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

, Contributors (2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

38.

Mistridis

, Krumm

, Monsch

, Berres

, Taylor

(2015) The 12 years preceding mild cognitive impairment due to Alzheimer’s disease: The temporal emergence of cognitive decline. J Alzheimers Dis 48, 1095–1107.

39.

Amieva

, Le Goff

, Millet

, Orgogozo

, Pérès

, Barberger-Gateau

, Jacqmin-Gadda

, Dartigues

(2008) Prodromal Alzheimer’s disease: Successive emergence of the clinical symptoms. Ann Neurol 64, 492–498.

40.

About NACC data, The UDS study population.

41.

Mori

, Yoneda

, Yamashita

, Hirono

, Ikeda

, Yamadori

(1997) Medial temporal structures relate to memory impairment in Alzheimer’s disease: An MRI volumetric study. J Neurol Neurosurg Psychiatry 63, 214–221.

42.

Masur

, Sliwinski

, Lipton

, Blau

, Crystal

(1994) Neuropsychological prediction of dementia and the absence of dementia in healthy elderly persons. Neurology 44, 1427–1432.

43.

Rabin

, Wang

, Katz

, Derby

, Buschke

, Lipton

(2012) Predicting Alzheimer’s disease: Neuropsychological tests, self-reports, and informant reports of cognitive difficulties. J Am Geriatr Soc 60, 1128–1134.

44.

Tierney

, Moineddin

, McDowell

(2011) Prediction of all-cause dementia using neuropsychological tests within 10 and 5 years of diagnosis in a community-based sample. J Alzheimers Dis 22, 1231–1240.

45.

van de Pol

, Korf

ESC

, van der Flier

, Brashear

, Fox

, Barkhof

, Scheltens

(2007) Magnetic resonance imaging predictors of cognition in mild cognitive impairment. Arch Neurol 64, 1023.

46.

Bora

, Yücel

, Pantelis

(2010) Cognitive impairment in schizophrenia and affective psychoses: Implications for DSM-V criteria and beyond. Schizophr Bull 36, 36–42.

47.

Stern

(2009) Cognitive reserve. Neuropsychologia 47, 2015–2028.

48.

Strydom

, Chan

, Fenton

, Jamieson-Craig

, Livingston

, Hassiotis

(2013) Validity of criteria for dementia in older people with intellectual disability. Am J Geriatr Psychiatry 21, 279–288.

49.

Kessing

L V

, Andersen

(2004) Does the risk of developing dementia increase with the number of episodes in patients with depressive disorder and in patients with bipolar disorder? J Neurol Neurosurg Psychiatry 75, 1662–1666.

50.

Forlenza

O V

, Aprahamian

, Radanovic

, Talib

, Camargo

, Stella

, Machado-Vieira

, Gattaz

(2016) Cognitive impairment in late-life bipolar disorder is not associated with Alzheimer’s disease pathological signature in the cerebrospinal fluid. Bipolar Disord 18, 63–70.

51.

Rise

, Haro

, Gjervan

(2016) Clinical features, comorbidity and cognitive impairment in elderly bipolar patients. Neuropsychiatr Dis Treat 12, 1203–1213.