Saccadic eye movement changes in Parkinson's disease dementia and dementia with Lewy bodies

Abstract

Neurodegeneration in Parkinson's disease dementia (PDD) and dementia with Lewy bodies (DLB) affect cortical and subcortical networks involved in saccade generation. We therefore expected impairments in saccade performance in both disorders. In order to improve the pathophysiological understanding and to investigate the usefulness of saccades for differential diagnosis, saccades were tested in age- and education-matched patients with PDD (n = 20) and DLB (n = 20), Alzheimer's disease (n = 22) and Parkinson's disease (n = 24), and controls (n = 24). Reflexive (gap, overlap) and complex saccades (prediction, decision and antisaccade) were tested with electro-oculography. PDD and DLB patients had similar impairment in all tasks (P > 0.05, not significant). Compared with controls, they were impaired in both reflexive saccade execution (gap and overlap latencies, P < 0.0001; gains, P < 0.004) and complex saccade performance (target prediction, P < 0.0001; error decisions, P < 0.003; error antisaccades: P < 0.0001). Patients with Alzheimer's disease were only impaired in complex saccade performance (Alzheimer's disease versus controls, target prediction P < 0.001, error decisions P < 0.0001, error antisaccades P < 0.0001), but not reflexive saccade execution (for all, P > 0.05). Patients with Parkinson's disease had, compared with controls, similar complex saccade performance (for all, P > 0.05) and only minimal impairment in reflexive tasks, i.e. hypometric gain in the gap task (P = 0.04). Impaired saccade execution in reflexive tasks allowed discrimination between DLB versus Alzheimer's disease (sensitivity > or =60%, specificity > or =77%) and between PDD versus Parkinson's disease (sensitivity > or =60%, specificity > or =88%) when +/-1.5 standard deviations was used for group discrimination. We conclude that impairments in reflexive saccades may be helpful for differential diagnosis and are minimal when either cortical (Alzheimer's disease) or nigrostriatal neurodegeneration (Parkinson's disease) exists solely; however, they become prominent with combined cortical and subcortical neurodegeneration in PDD and DLB. The similarities in saccade performance in PDD and DLB underline the overlap between these conditions and underscore differences from Alzheimer's disease and Parkinson's disease.

Saccadic latency distributions in Parkinson’s disease and the effects of l-dopa

“…Given the intersubject variability and degree of overlap of saccadic latency distributions from PD patients and controls it is clear that as it currently stands this technique could not be used as a diagnostic biomarker of PD-a conclusion compatible with other recent studies involving nondemented PD patients, although these were analysed differently (Mosimann et al 2005). Nevertheless, saccadic latency does seem to correspond to severity of disease (Fig.…”

“…Nevertheless, saccadic latency does seem to correspond to severity of disease (Fig. 5) and the presence of dementia (Mosimann et al 2005), suggesting that it might prove to be a useful marker of disease progression-perhaps in particular of cognitive decline. Clearly this requires investigation in longitudinal studies, but recently simple reaction and movement times have shown promise as biomarkers by detecting changes in motor performance over time in patients undergoing embryonic nigral cell transplantation, and correlating with UPDRS 'off' scores at 4 and 12 months of post-transplantation (Gordon et al 2004).…”

“…Given the heterogeneity of PD it is likely that we will require different biomarkers to track the development of different cognitive, motor or pathological parameters as they progress at different rates over time. Since 'reflexive' saccadic latency seems to be most prolonged by a combination of cortical and subcortical pathology rather than either in isolation (Mosimann et al 2005), one might speculate that analysis of saccadic latency distributions will turn out to be a particularly sensitive biomarker of the onset and progression of cognitive defects and dementia within a PD population (Foltynie et al 2004a), and might enable the disease progression to be described according to the LATER decision model.…”

“…Saccadic eye movements have quite frequently been investigated as a potential biomarker of PD since they have a highly reproducible trajectory (Roy-Byrne et al 1995), and there is evidence that they might be affected by cognitive decline, depression or other aspects of the disease (Leigh and Kennard 2004;Mosimann et al 2005;Briand et al 1999;Hikosaka et al 2000). However, these studies are not easy to compare given the different target populations (early or advanced disease, medication effects), testing paradigms used (gap or overlap of stimuli, 'reflexive' or 'voluntary' task) and parameters of the saccade measured (velocity, amplitude, duration, errors).…”

Distinctive features of microsaccades in Alzheimer’s disease and in mild cognitive impairment

“…Antisaccades (i.e., volitional saccades with opposite direction to the target) in AD exhibit impaired inhibition towards the target, as well as impaired correction of errors, with the extent of the deficiencies being related to the severity of the disease (Hershey et al 1983;Fletcher and Sharpe 1986;Moser et al 1995;Shafiq-Antonacci et al 2003;Crawford et al 2005;Garbutt et al 2008). Prosaccades (i.e., saccades directed toward the target) in AD have abnormally long latencies (Hershey et al 1983;Fletcher and Sharpe 1986;Moser et al 1995;Shafiq-Antonacci et al 2003;Crawford et al 2005;Garbutt et al 2008), but see Hershey et al (1983) and Mosimann et al (2005). Saccadic gain and speed findings in AD are controversial: some studies found impairment (Fletcher and Sharpe 1986;Hotson and Steinke 1988;Shafiq-Antonacci et al 2003) whereas others did not (Moser et al 1995;Garbutt et al 2008).…”