The influence of two common dementia types on visual symptoms


Dr. Joanna M. Jefferis
NIHR Doctoral Research Fellow
Clinical Ageing and Research Unit
Campus for Ageing and Vitality
Newcastle Upon Tyne NE4 5PL
Tel.: 01912481297
Fax: 01912481251


Dementia with Lewy bodies (DLB) and Alzheimer’s disease (AD) represent two common forms of dementia in older people. Compared with AD, people with DLB have less memory impairment but more visuospatial impairment, and they are more likely to experience persistent, well-formed visual hallucinations (Mosimann et al. 2004 and McKeith et al. 2005). DLB also shows considerable overlap with Parkinson’s disease dementia that has recently been reported to have a broad range and high burden of visual symptoms (Archibald et al. 2011). We aimed to determine the influence of DLB and AD on visual symptoms relevant to eye care providers.

We recruited 21 people with AD and 21 with DLB from old-age psychiatry services and 19 healthy controls. Diagnoses of AD and DLB were made by two independent, expert assessors according to established criteria (McKhann et al. 1984 and McKeith et al. 2005). Participants were assessed in their home with a relative or carer present. Visual symptoms were assessed using a simple questionnaire designed for people with cognitive impairment. The questionnaire, which has previously been used in a group of people with Parkinson’s disease dementia (Archibald et al. 2011), was interviewer administered and sought to be of low burden to people with dementia. It contained a small number of short questions that were asked in the presence of an informant who could corroborate the answers. It enquired about the presence of six visual symptoms (blurred vision, double vision, misjudging objects, difficulty reading, painful/dry eyes and photosensitivity) and their frequency (never, less than once a week, once a week, several times a week, once a day and several times a day). Frequency of symptoms was scored from 0 (never) to 5 (several times a day). Visual hallucinations were screened for using the North East Visual Hallucinations Inventory (NEVHI) (Mosimann et al. 2008). Visual acuity was measured on a decimal scale using best correction with a standardized Snellen equivalent near visual acuity chart at 40 cm. Cognition was assessed with the mini-mental state examination (MMSE).

Subjects were of a similar age (mean age ± SD: AD 82.7 ± 8.2; DLB 80.7 ± 5.9; controls 77.6 ± 7.1; one-way anova, p = 0.092) and had similar levels of visual acuity (mean decimal acuity ± SD: AD 0.65 ± 0.23; DLB 0.63 ± 0.22; controls 0.67 ± 0.25; one-way anova, p = 0.913). The mean MMSE ± SD was 19.9 ± 4.2 for those with AD, 19.1 ± 5.4 for those with DLB and 29.0 ± 1.2 for controls. These MMSE differences were significant across the three groups (one-way anova, p < 0.001), but not significant between AD and DLB participants (t-test, p = 0.60).

Not surprisingly, all subjects with DLB experienced visual hallucinations (21/21) compared with a minority of AD participants (3/21) and none of the controls (χ2, p < 0.001). However, DLB subjects also had a significantly greater burden of other visual symptoms, something not previously recognized (Table 1).

Table 1.   Visual symptoms and their frequency for those with Lewy body dementia (DLB), Alzheimer’s disease (AD) and controls. The frequency score is a score from 0 to 5 where 0 refers to never and 5 refers to several times a day.
SymptomDLB (n = 21)AD (n = 21)Control (n = 19)p-valueDLB (n = 21)AD (n = 21)Control (n = 19)p-value
Number experiencing symptom (%)Median Frequency Score (IQR)
  1. IQR = interquartile range.

  2. * p-Values <0.05 are considered significant.

  3. Chi-squared or Fisher’s exact test as appropriate.

  4. Mann–Whitney U test.

  5. a DLB vs. AD; bDLB vs. control; cControl vs. AD.

Blurred vision10 (48)1 (5)4 (21) a0.002*, b0.079, c0.1720 (0–3)0 (0–0)0 (0–0) a0.005*, b0.295, c0.12
Double vision10 (48)0 (0)3 (16) a<0.001*, b0.032*, c0.0980 (0–3.5)0 (0–0)0 (0–0) a<0.001*, b0.071, c0.058
Misjudging objects13 (62)0 (0)1 (5) a<0.001*, b<0.001*, c0.4753 (3–5)0 (0–0)0 (0–0) a<0.001*, b0.001*, c0.287
Difficulty reading10 (48)1 (5)1 (5) a0.002*, b0.003*, c1.00 (0–4.5)0 (0–0)0 (0–0) a0.002*, b0.008*, c0.942
Photosensitivity5 (24)2 (10)4 (21) a0.41, b1.0, c0.3980 (0–2)0 (0–0)0 (0–0) a0.214, b0.895, c0.308
Painful/dry eyes11 (52)0 (0)2 (11) a<0.001*, b0.003*, c0.2191 (1–3.75)0 (0–0)0 (0–0) a<0.001*, b0.009*, c0.127

People with DLB have significantly more visual complaints when compared to those with AD and controls. The difficulty reported here with misjudging objects and reading in DLB is in keeping with the visuoperceptual difficulties they experience (Mosimann et al. 2004). Some of the other difficulties reported may be due to parkinsonian features that commonly occur in DLB and have previously been reported to increase double vision (Archibald et al. 2011) and dry eye disorders (Tamer et al. 2005) although why this should be the case remains to be established. Interestingly, those with AD reported less visual symptoms than controls (although not significantly so). This may be due to under-reporting in the AD group, or anosognosia (an unawareness of symptoms) that is not uncommon in AD.

We have asked about a broad range of visual symptoms in a way that is understandable for patients with dementia. The study is not without its limitations; the sample size is small, and there was no validated measure of visual symptoms in those with dementia and as such we used a nonvalidated questionnaire. Visual acuities were measured only at near, in order that assessments could be carried out in participants' own residence.

In summary, as well as visual hallucinations that are a recognized core feature, those with DLB also experience a broad range of visual symptoms with which they may present to eye care providers, and which may be misattributed to coexisting age-related ocular changes, such as cataract. The potential role visual symptoms may play in the early recognition, and diagnosis of DLB deserves further study.