Article Text

Download PDFPDF

Research paper
Psychosis and longitudinal outcomes in Huntington disease: the COHORT Study
  1. Michael H Connors1,2,3,
  2. Armando Teixeira-Pinto4,
  3. Clement T Loy4,5,6
  1. 1Sydney Medical School, The University of Sydney, Sydney, New South Wales, Australia
  2. 2Centre for Healthy Brain Ageing, The University of New South Wales, Sydney, New South Wales, Australia
  3. 3Dementia Centre for Research Collaboration, The University of New South Wales, Sydney, New South Wales, Australia
  4. 4Sydney School of Public Health, The University of Sydney, Sydney, New South Wales, Australia
  5. 5Brain and Mind Centre, The University of Sydney, Sydney, New South Wales, Australia
  6. 6The Garvan Institute of Medical Research, Sydney, New South Wales, Australia
  1. Correspondence to Professor Clement T Loy, Brain and Mind Centre, University of Sydney, Sydney, NSW 2006, Australia; clement.loy{at}sydney.edu.au

Abstract

Objective Huntington disease (HD) is an autosomal dominant neurodegenerative disease involving motor disturbances, cognitive decline and psychiatric symptoms. Psychotic symptoms occur in a significant proportion of patients. We sought to characterise the clinical outcomes of this group of patients.

Methods Data were drawn from the Cooperative Huntington Observational Research Trial, a prospective, multi-centre observational study. 1082 patients with HD were recruited. Measures of cognition, function, behavioural disturbance and motor function were completed annually over 5 years.

Results Overall, 190 patients (17.6%) displayed psychotic symptoms. These patients demonstrated worse cognition, function and behavioural disturbances than patients without psychosis over time. Patients with psychosis also demonstrated lower levels of chorea than patients without psychosis, despite adjusting for concurrent antipsychotic and tetrabenazine use.

Conclusions Psychosis in HD is associated with poorer outcomes in cognition, function and behavioural symptoms. Patients with psychotic symptoms may also have less chorea. Altogether, the findings suggest patients with psychosis have a distinct clinical course.

  • chorea
  • Huntington disease
  • longitudinal
  • psychosis
  • prognosis
View Full Text

Statistics from Altmetric.com

Introduction

Huntington disease (HD) is an autosomal dominant neurodegenerative disease that is characterised by motor disturbances, cognitive impairment and psychiatric symptoms.1 2 Onset is typically in mid-life, though can occur at any age, and motor disturbances, particularly chorea (involuntary, jerky movements), are often the first signs of the disease.3 The disease results from an unstable expansion of the cytosine–adenine–guanine (CAG) repeat in the Huntingtin gene4 and the number of these repeats correlates with the disease’s onset5 and progression.6 HD occurs in approximately one in 10 000 people7 and there is no known prevention or cure. Given its relative rarity, there are limited data on whether features in its clinical presentation predict subsequent disease progression.8

Psychotic symptoms occur in 3%–11% of patients9–11 and it remains unclear whether these patients experience a different clinical course to those without such symptoms (for simplicity, we refer to psychosis—delusions and hallucinations—as symptoms, though they may also be considered to be signs). In other neurodegenerative disorders, psychosis is associated with a more rapid decline in cognition and function over time.12–15 In Alzheimer’s disease and frontotemporal dementia, there is also evidence for distinct genetic underpinnings of psychotic phenotypes.16–20 It is unclear if similar findings hold in HD. In cross-sectional analyses, psychotic symptoms in HD have been associated with lower functional ability, greater motor disturbance, longer duration of illness9 and lower cognition.21 These findings, however, are limited by the relatively small number of patients with psychotic symptoms in the respective samples and longitudinal outcomes remain unknown. In addition, psychotic features in HD have been found to aggregate in families22–25 and appear to be unrelated to the number of CAG repeats in the Huntingtin gene.9 25–29 These latter findings are consistent with the possibility of a psychotic endophenotype with a distinct genetic basis, though specific genetic modifiers have yet to be identified.

Against this background, we attempted to characterise the clinical features of the psychotic phenotype in HD. We examined the longitudinal outcomes of a large sample of patients with HD and compared those with psychotic symptoms to those without psychotic symptoms. We drew on data from the Cooperative Huntington Observational Research Trial (COHORT),30 which assessed cognitive ability, functional ability, motor disturbances and behavioural symptoms at baseline and over the following 5 years. We considered chorea and other motor disturbances—such as dystonia (involuntary muscle co-contractions that produce abnormal posturing), rigidity and balance—separately given evidence of distinct neural underpinnings and different trajectories over time.2 8 31 We hypothesised that psychosis would be associated with poorer cognitive, functional and motor abilities, and worse behavioural symptoms.

Methods

Design

Participants were from the COHORT study,30 a prospective, multi-centre observational study involving 44 separate testing sites across Australia (n=2), Canada (n=4) and the USA (n=38). The study recruited four groups of participants: (1) individuals with a clinical diagnosis of HD; (2) individuals at high risk of HD according to DNA testing but who did not have clinically diagnosed HD; (3) first-degree and second-degree relatives of individuals from the first two groups and (4) spouses or caregivers of individuals from the first two groups (controls) who had no genetic risk for HD. The study collected blood for genetic analyses at enrolment and behavioural measures annually for up to 5 years. All participants provided written informed consent. The current paper represents a retrospective analysis of the study data; it was conceived after data collection was complete, so there was no risk of recall bias.

Participants

Our analyses focused on patients with a diagnosis of HD at baseline or who received a diagnosis during the course of the study. This was based on the motor subscale of the Unified Huntington disease Rating Scale (UHDRS),32 which requires the clinician to rate whether they believe with a confidence level of ≥99% that the subject has manifest HD. In addition, all patients were required to have a CAG count greater than 35. Psychosis was defined by either the presence of psychotic symptoms at that time point (according to the UHDRS behavioural assessment, items 34 and 35) or a previous history of psychotic symptoms (regardless of when this occurred relative to patients’ diagnosis of HD).

Measures and procedure

At baseline, participants completed a medical history and gave a blood sample for DNA testing and Huntingin CAG repeat genotyping. At baseline and each visit, participants completed a detailed physical and neurological examination. Motor disturbance was assessed using the UHDRS ’99,32 which provides a total score of motor disturbance (range: 0–124, higher scores indicate more severe symptoms). The subscores for chorea (range: 0–28), dystonia (range: 0–20), rigidity (range: 0–8) and balance (combining scores for gait, tandem gait and postural instability; range: 0–12) were also examined. Cognition was assessed using the Mini-Mental State Examination33 (MMSE, range: 0–30; higher scores indicate better cognition). Function was assessed using the UHDRS measures of total functional capacity (TFC; range: 0–13; higher scores indicate better function) and independence (range: 0–100; higher scores indicate better function). Behavioural disturbance was assessed using the UHDRS total score of behavioural disturbance excluding delusions and hallucinations (range: 0–144; higher scores indicate more severe and frequent behavioural disturbances). A list of patients’ medications, including antipsychotics (typically prescribed for both psychosis and chorea) and tetrabenazine (typically prescribed for chorea), was also compiled.34

Statistical analyses

The baseline characteristics of all patients—including both those with HD at baseline and those subsequently diagnosed—were compared according to whether they ever experienced psychosis (both prior to baseline or during the study). Data were compared using logistic regression to control for time since clinical diagnosis. The relationship between number of CAG repeats and age of onset of psychotic symptoms was examined using Spearman’s rank-order correlation coefficient.

Longitudinal data were analysed using linear mixed models with normally distributed random intercepts and random effects for time. We restricted the analyses to the first 4 years of data due to the relatively small number of patients returning for the 5-year follow-up (see online supplementary appendix 1). Only patients with a current motor diagnosis of HD were included in the analyses (patients diagnosed with HD during the study were only included once they had received their diagnosis). Psychosis was defined as a time-varying variable, such that patients were categorised as having psychosis once they showed active symptoms or if they had a previous history of psychosis. Outcome measures were cognition (MMSE), function (TFC), motor disturbances (total score, chorea, dystonia, rigidity and balance) and behavioural disturbances (total behavioural score excluding delusions and hallucinations). Time was measured in years from the patients’ clinical diagnosis.

For each outcome, separate models included the following predictors: age, sex, number of CAG repeats and the presence of psychosis. For behavioural disturbances, use of antipsychotic medications was included as a dichotomous time-varying variable. Likewise, for motor disturbances, use of antipsychotic medication and tetrabenazine were included as dichotomous time-varying variables. For all outcomes, interactions between psychosis and time were included in the model to check if the effect varied over time and were retained in the model if p<0.10. If statistically significant interactions involving a categorical variable were statistically significant, separate effect estimates of the interaction were reported for each value of the categorical variable. Models were selected on the basis of the Akaike information criterion. Statistical significance was set at p<0.05 for all statistical tests of main effects given the exploratory nature of the analyses.

Two sensitivity analyses were conducted. First, to examine the possibility that differences between patients with psychosis and those without psychosis represent long-standing phenotypic traits that antedate overt psychotic symptoms, analyses were repeated defining psychosis as time-invariant (including all patients who had a history of psychosis or displayed psychotic symptoms at any point over the study). Second, to examine the contribution of active psychotic symptoms to the findings, analyses were repeated distinguishing active psychotic symptoms as a time-variant variable from both previous psychosis and no previous psychosis. All analyses were completed using SPSS V.25 (IBM Corporation, Armonk, New York, USA).

Results

Over the course of the study, 1082 patients with HD were recruited (994 patients had HD at baseline; a further 88 were diagnosed with HD during the study). Of these, 190 (17.6%) demonstrated psychotic symptoms at some point during the course of their disease (97 patients had a history of psychosis before the study and 141 displayed psychotic symptoms during the study; 48 displayed such symptoms both before and during the study). Of the patients who developed psychotic symptoms during the study, 121 (85.8%) developed delusions and 39 (27.7%) developed hallucinations; 19 (13.5%) developed both symptoms.

The characteristics of patients at enrolment—including both those with HD at the baseline visit and those subsequently diagnosed—are shown in table 1. Patients with either active psychosis or a past history of psychosis did not differ from those without psychosis in terms of age, sex, education or CAG repeats. Patients with psychosis, however, exhibited worse function, cognition and behavioural symptoms than patients without psychosis. Patients with psychosis also had less chorea than patients without psychosis. The sample size across the study is shown in the supplementary material (see online supplementary figure 1).

Table 1

Characteristics of patients at baseline (including both patients with Huntington disease at baseline and patients who were subsequently diagnosed during the study)

Data about age of onset of psychotic symptoms were available for 138 (72.6%) patients. For these patients, the mean age of onset of psychotic symptoms was 51.38 years (SD=12.29). The mean time between clinical diagnosis and onset of psychotic symptoms was 3.58 years (SD=4.81); 21 patients (15.2%) exhibited psychotic symptoms prior to their HD diagnosis; 21 (15.2%) had psychotic symptoms at the time of their HD diagnosis and 96 (69.6%) developed psychotic symptoms after their HD diagnosis. There was a negative correlation between number of CAG repeats and age of onset of psychosis, rs(136) = −0.66, p<0.001.

In linear mixed model analyses, patients with psychosis demonstrated worse cognition, function and behavioural disturbances than patients without psychosis. For measures of function, there was a significant interaction between psychosis and time using TFC, but not the independence scale (table 2). In particular, patients with psychosis had TFC scores 1.18 lower (p<0.001) on average than patients without psychosis at the time of diagnosis in the model, but declined at a slower rate thereafter, adjusting for patient’s age, sex and CAG repeats. Patients with psychosis had independence scale scores 4.97 lower (p<0.001) than patients without psychosis across the study on average adjusting for other covariates.

Table 2

Linear mixed models for measures of function

For cognition, patients with psychosis had MMSE scores 1.26 units lower on average than patients without psychosis (p<0.001) when adjusting for other variables (table 3). For behavioural symptoms, patients with psychosis had behavioural scores 8.55 units higher on average than patients without psychosis (p<0.001) when adjusting for other variables (table 3). There were no interactions with time for either cognition or behavioural symptoms, indicating the differences between patients with psychosis and patients without psychosis were stable over time.

Table 3

Linear mixed models for cognition and behavioural disturbances

For movement disturbances, results varied according to the type of disturbance. Patients with psychosis demonstrated less chorea than patients without psychosis (p=0.001), scoring 1.08 units lower on average after adjusting for age, sex, CAG repeats, antipsychotic medication and time since onset of symptoms (table 4). Patients with psychosis also demonstrated worse balance than patients without psychosis (p=0.025), scoring 0.32 units higher on average after adjusting for other variables (table 4). There were no significant differences between patients with psychosis and patients without psychosis in dystonia, rigidity and overall motor score (table 4). There were no interactions with time for any of the motor score variables.

Table 4

Linear mixed models for motor disturbances

The effects for all outcome variables were unchanged when analyses were based on only psychotic symptoms that occurred during the study or when examining delusions and hallucinations separately (results not shown).

A sensitivity analysis found similar findings when treating psychosis as a time-invariant variable—such that all participants who developed psychotic symptoms during the study or who had a previous history of psychosis were considered together as a stable group, even if this preceded the onset of their psychotic symptoms (see online supplementary tables 1 and 2). In particular, patients with psychosis defined in this way displayed lower function (both TFC and independence), lower cognition and greater behavioural disturbances than patients without psychosis. These patients also developed chorea at a slower rate than patients without psychosis, adjusting for antipsychotic and tetrabenazine use and other covariates.

A further sensitivity analysis distinguished patients with active psychotic symptoms at any time point from patients with previous psychotic symptoms and patients who had never had psychotic symptoms. Both groups of patients with psychotic symptoms—those with active symptoms and those who had previous symptoms—demonstrated lower function (both TFC and independence), lower cognition and greater behavioural disturbances than patients who had never experienced psychosis. Patients with active psychosis demonstrated worse rigidity and balance than patients who had never experienced psychosis. Patients with previous psychosis also demonstrated less chorea than patients who had never experienced psychosis (see online supplementary tables 1 and 2).

Discussion

A significant proportion of patients with HD—approximately 18%—exhibited psychotic symptoms. These patients demonstrated lower cognitive ability, lower functional abilities and greater behavioural disturbances compared with patients without psychosis, and showed a different trajectory in function over time. Psychosis was also associated with a different profile of motor disturbances to patients without psychosis. In particular, patients with psychosis had less chorea and worse balance than patients without psychosis, despite adjustment for antipsychotic and tetrabenazine use. These differences in motor function remained stable over time and the lower levels of chorea were apparent regardless whether psychotic symptoms were actively present, suggesting the presence of a long-standing trait. Altogether, the current findings indicate that patients with HD with psychosis have a distinct clinical course.

The proportion of patients with psychotic symptoms was higher than previous cross-sectional studies.9–11 The higher proportion may be due partly to the lower threshold for psychosis used in our analyses (we considered the presence of symptoms, rather than only clinically significant symptoms), the regular assessment of patients within the longitudinal design of the study, and the recruitment of patients from HD specialists, which could lead to more patients in the sample with challenging or unusual presentations. Nevertheless, the finding that psychosis is associated with cognitive and functional deficits is consistent with previous research, which has found that patients with psychosis have lower cognition21 and function9 in cross-sectional comparisons. The current study extends previous research by showing these patients differ in terms of other behavioural symptoms and such various differences exist longitudinally and after controlling for demographic and clinical variables. The current study also reveals that while the cognitive deficits associated with psychosis may be stable over time, functional deficits—particularly when measured using TFC—may progress more slowly in patients with psychosis than patients without psychosis. These latter findings may be the result of a floor effect: Patients with psychosis may reach significant levels of impairment earlier in the course of their disease and thereafter deteriorate more slowly. Alternatively, these findings may reflect slower neurodegeneration and disease progression in patients with psychosis.

By contrast, the finding that psychosis is associated with a different profile of motor disturbances is less expected. This is particularly the case given its dissociation from poorer cognitive and functional outcomes and balance. Other research9 has found that patients with psychosis have greater motor disturbances overall, though did not distinguish types of motor disturbances or control for antipsychotic medication and tetrabenazine use, and was limited by the small number of patients with psychosis in their sample. The lower levels of chorea in patients with psychosis that was found in the current study have not been previously reported. The current study further reveals that these differences in motor function are stable over time and not due to active psychosis or the presence or absence of medication. Possible mechanism are unclear, though altered dopaminergic activity has been implicated in both psychosis35 and chorea.36 As far as we are aware, there have been no detailed neuroimaging or neuropathological studies of psychosis in HD.

The current study had a number of limitations. First, the study was limited by its convenience sampling of patients and the fact that patients were volunteers recruited from HD specialists. As such, patients may have come from a higher socio-economic background and differ in the severity of their disease and rates of psychosis to those seen in other clinical settings. Second, data were available for 4 years, which is only a small proportion of the course of HD, so longer-term trajectories are unclear. Third, it was not possible to determine the duration of psychotic episodes. Finally, analysis of medications was limited and lacked a fine-grained approach to dosage and duration. While there are a number of competing methods for standardising antipsychotic doses in terms of their effect on psychotic symptoms in schizophrenia,37 antipsychotics have different motor side effect profiles and there is no established way of standardising doses in terms of their effect on chorea or in a HD population. This is significant given that a large proportion of patients without psychosis received antipsychotic medications during the study, presumably for their chorea. It is thus possible that patients with psychosis received higher doses of antipsychotics than patients without psychosis.

Despite these limitations, the current study provides evidence that patients with psychosis have a distinct clinical course characterised by more severe cognitive, functional and behavioural deficits, yet also relatively less chorea. The findings, if replicated in other cohorts, point to an important direction for future research in terms of clarifying the genetic and neurobiological underpinnings of this psychotic endophenotype.

Acknowledgments

We thank the Huntington Study Group COHORT investigators and coordinators, who collected data and/or samples for this study, as well as all participants and their families, who made this work possible.

References

View Abstract

Footnotes

  • Presented at An earlier version of this work was presented at the Neuropsychiatric Syndromes symposium of the International Society to Advance Alzheimer's Research and Treatment (ISTAART) Professional Interest Area day at the Alzheimer’s Association International Conference (AAIC) on 13 July 2019 in Los Angeles, USA.

  • Contributors MHC helped to conceptualise the paper, analysed the data, interpreted the data and drafted the manuscript. AT-P helped to conceptualise the paper, provided statistical advice, interpreted the data and revised the manuscript for intellectual content. CTL contributed to acquisition of the data, helped to conceptualise the paper, provided statistical advice, interpreted the data and revised the manuscript for intellectual content.

  • Funding MHC was partially supported by the Dementia Centre for Research Collaboration, which is funded by the National Health and Medical Research Council (NHMRC) in Australia. AT-P was partially supported by NHMRC program grant 633003 to the Screening & Test Evaluation Program (STEP). CTL was supported by a NHMRC Dementia Research Development Fellowship (APP1107657). The CHDI Foundation, a non-for-profit organisation dedicated to finding treatments for people with Huntington disease, funded the COHORT study. The CHDI Foundation had no role in the analyses, data interpretation or writing of this paper.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval Ethics approval was obtained from institutional ethics committees associated with individual testing centres (National Institute of Health clinical trials registry number: NCT00313495).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.