Neuropsychiatric symptoms in sporadic Creutzfeldt-Jakob disease

Abstract

Although neuropsychiatric symptoms are not part of diagnostic criteria for sporadic Creutzfeldt-Jakob disease (sCJD), a few retrospective studies and our clinical experience have suggested they are prominent and often occur early. The objective of our study was to assess prospectively the neuropsychiatric features in sCJD (and their impact on caregivers) and compare them with five other neurodegenerative diseases: Alzheimer's disease (AD), dementia with Lewy bodies (DLB), progressive supranuclear palsy, behavioural variant frontotemporal dementia (bvFTD) and semantic variant primary progressive aphasia. The Neuropsychiatric Inventory (NPI) was given at the first UCSF research visit to caregivers of 789 serial patients evaluated at the UCSF Memory and Aging Center from January 2000 through November 2016 who met the diagnostic research criteria for one of these six neurodegenerative disorders. All subjects underwent a Mini-Mental State Examination, and demographic data (age, sex, years of education) were collected. We hypothesized that sCJD has a very prominent behavioural and neuropsychiatric profile, which might help to distinguish it from other neurodegenerative diseases. Of our sCJD cohort (n = 92), 97% exhibited at least one and 50% at least six of the 12 symptoms on the NPI at the time of their first research visit. The most common behavioural symptoms, which occurred in >50% of the sCJD patient cohort, were: appetite/eating disturbances (68%), apathy/indifference (66%), night-time behaviour disturbances (53%), aberrant motor behaviour (53%) and anxiety (52%). Even the least common behaviour, disinhibition, occurred in 19%. Compared with DLB and bvFTD, two conditions with very prominent behavioural features that are included in their diagnostic criteria, sCJD had significantly higher mean NPI frequency × severity scores for night-time behaviour disturbances and delusions. Patients with sCJD scored higher for hallucinations and dysphoria than bvFTD patients and higher for appetite/eating disturbances, aberrant motor behaviour and agitation than patients with DLB. sCJD resulted in significantly worse frequency × severity scores than AD in nine of the NPI symptom domains, with our data revealing sCJD to be a highly behavioural syndrome. No neuropsychiatric symptom is pathognomonic for sCJD, but specific symptoms may enable the differentiation of sCJD from other neurodegenerative diseases. Our findings support the inclusion of behavioural symptoms in sCJD diagnostic criteria.

Keywords: CJD; Jakob-Creutzfeldt disease; management; prion disease; transmissible spongiform encephalopathy.

Figure 1

Frequency and severity of Neuropsychiatric Inventory behaviours in sporadic Creutzfeldt-Jakob disease. This figure shows the percentage of sporadic Creutzfeldt-Jakob disease (sCJD) patients with each Neuropsychiatric Inventory (NPI) behaviour that occurred at a frequency of ‘often’ (at least ∼1×/week), as well as the distribution of severity for each symptom. Light grey bars indicate mild, darker grey bars indicate moderate and black bars indicate marked severity of the behaviour. For a description of the three categories of symptom severity, see Supplementary Table 2. NPI behaviours are presented in decreasing order of frequency. Note that the most frequently observed symptoms were not always the most severe, nor the most distressful to caregivers (Fig. 2). See Figs 3 and 4 for more qualitative descriptions and breakdowns of the sleep and appetite/eating behaviours, respectively.

Figure 2

Frequency and severity of caregiver distress in sporadic Creutzfeldt-Jakob disease. This figure shows the distribution of the severity of caregiver distress for each Neuropsychiatric Inventory (NPI) behaviour, with the behaviours presented in decreasing order of frequency of occurrence (of at least weekly, as in Fig. 1). For presentation purposes, some of the six caregiver distress severity categories were combined. Very light grey bars indicate no caregiver distress, light grey bars combine minimal and mild caregiver distress, grey bars indicate moderate and black bars indicate severe/extreme caregiver distress. Note that in general, with a few exceptions, the more frequent behaviours were associated with the highest percentages of moderate to severe/extreme caregiver distress (see main text). See Supplementary Table 2 for a description of caregiver distress severity categories. Note that some totals are one point different from the sum of the individual groups due to rounding the values for each severity category in the chart.

Figure 3

Qualitative breakdown of night-time behavioural disturbances in sporadic Creutzfeldt-Jakob disease. This diagram shows the percentages of night-time behavioural disturbances occurring ‘often’ (at least ∼1× /week). Note: all percentages in this figure refer to the entire cohort of 92 sporadic Creutzfeldt-Jakob disease (sCJD) subjects, not the subsets. Please see the main text for how the insomnia information was extracted from the Neuropsychiatric Inventory (NPI). This figure shows that, of the sCJD cohort, 53% had night-time behavioural disturbances, 46% had insomnia and 27% (5% + 22%) had daytime somnolence (with or without insomnia). Interpreted another way, 87% (46%/53%) of all patients with night-time behavioural disturbances had insomnia; of these, 48% (22%/46%) had insomnia with, and 52% (24%/46%) without, daytime somnolence. See Fig. 1 for a breakdown of the degrees of night-time behavioural disturbance severity and Fig. 2 for how these affect caregiver distress. Asterisk indicates the following questions: Does the patient wander, pace or engage in inappropriate activities at night, does the patient awaken you during the night, does the patient awaken at night dressed and plan to go out thinking it is morning, or other?

Figure 4

Prevalence of appetite/eating disturbances in sporadic Creutzfeldt-Jakob disease, with qualitative breakdown of weight and appetite changes. The Neuropsychiatric Inventory (NPI) appetite/eating disturbances category includes increased or decreased weight, increased or decreased appetite, change in eating behaviour and change in the kind of food he/she likes. Note: all percentages refer to the entire cohort of 92 sporadic Creutzfeldt-Jakob disease (sCJD) subjects, not the subsets. Of all sCJD subjects, 68% had appetite/eating disturbances, and more than half of these subjects (38%/68% = 56%) had weight loss. Of the entire sCJD cohort, 47% (15% + 32%) had no weight change, 15% had weight gain, and 38% had weight loss. Of the total sCJD cohort, 55% (9% + 3% + 11% + 32%) had no appetite change, 18% had increased appetite and 27% had decreased appetite. Interestingly, 37% (14%/38%) of those with weight loss did not have decreased appetite, perhaps suggesting a metabolic effect from the disease. ^aThe percentages in this figure refer only to weight and appetite changes and not to other appetite/eating disturbances that subjects might have had, including (i) putting too much food in mouth; (ii) change in the kind of food they like; (iii) eating the same food each day or eating food in exactly the same order; or (iv) other changes in eating/appetite. See ‘Results’ and ‘Discussion’ sections for a comparison between food preference changes in sCJD and behavioural variant frontotemporal dementia.

Figure 5

Comparison of mean Neuropsychiatric Inventory frequency × severity scores in sporadic Creutzfeldt-Jakob disease versus other neurodegenerative diseases. The frequency × severity score (mean ± standard error) for non-Creutzfeldt-Jakob disease (non-CJD) cohorts, including Alzheimer’s disease (AD), dementia with Lewy bodies (DLB), progressive supranuclear palsy (PSP), behavioural variant frontotemporal dementia (bvFTD) and semantic variant primary progressive aphasia (svPPA) in the 12 Neuropsychiatric Inventory (NPI) categories. The behaviours are organized in decreasing order of their sporadic CJD (sCJD) frequency × severity scores. Asterisks denote categories for which the mean frequency × severity score in sCJD was statistically greater than for the specific non-CJD, and filled triangles highlight the opposite, with certain non-CJDs being more affected (P < 0.05). sCJD was associated with significantly worse delusions and hallucinations than all non-CJDs, apart from no difference from DLB. sCJD was worse than the prototypical behavioural non-CJD bvFTD in four NPI categories (night-time behavioural disturbances, depression/dysphoria, delusions and hallucinations). See Supplementary Table 5 for comparisons between sCJD and non-CJDs.