Cognitive dysfunction in rapid eye movement sleep behavior disorder


  • Josie-Anne Bertrand,

    1. Center for Advanced Research in Sleep Medicine, Hôpital du Sacré-Cœur de Montréal
    2. Department of Psychology, Université de Montréal
    Search for more papers by this author
  • Daphné Génier Marchand,

    1. Center for Advanced Research in Sleep Medicine, Hôpital du Sacré-Cœur de Montréal
    2. Department of Psychology, Université du Québec à Montréal
    Search for more papers by this author
  • Ronald B Postuma,

    1. Center for Advanced Research in Sleep Medicine, Hôpital du Sacré-Cœur de Montréal
    2. Department of Neurology, Montreal General Hospital, Montreal, Quebec, Canada
    Search for more papers by this author
  • Jean-François Gagnon

    Corresponding author
    1. Center for Advanced Research in Sleep Medicine, Hôpital du Sacré-Cœur de Montréal
    2. Department of Psychology, Université du Québec à Montréal
    Search for more papers by this author

Dr Jean-François Gagnon, Centre d'Études Avancées en Médecine du Sommeil, Hôpital du Sacré-Cœur de Montréal, 5400 boul. Gouin ouest, Montréal, Québec, Canada, H4J 1C5. Email:


Cognitive impairment is a frequent feature of rapid eye movement sleep behavior disorder (RBD). The cognitive profile of RBD patients is heterogeneous, with impairments in attention, executive functions, episodic memory, and visuospatial abilities. Moreover, over 50% of RBD patients meet the diagnostic criteria for mild cognitive impairment (MCI). Although a comprehensive neuropsychological assessment remains the most sensitive way to detect MCI, three cognitive screening tests have been validated in RBD. The Montreal Cognitive Assessment was found to be the most appropriate screening test for detecting MCI in RBD. In addition, RBD in Parkinson's disease may be a risk factor for MCI and dementia.

Rapid eye movement sleep behavior disorder (RBD) is a parasomnia that is considered a risk factor for synucleinopathies, a category of neurodegenerative diseases including Parkinson's disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy.1 Several markers of synucleinopathies have been identified in idiopathic RBD.2–4 This review focuses on cognitive impairment in RBD.


Although many cognitive domains have been defined, the main ones are learning and memory, language, attention, executive functions, visuospatial abilities, and praxis. Attention and executive functions are often considered together. A significant proportion of idiopathic RBD patients complain about reduced cognitive abilities.5 To our knowledge, six controlled studies have assessed cognitive performance in idiopathic RBD patients.5–10Table 1 presents the sociodemographic and neuropsychological variables. To facilitate between-study comparison on cognitive tasks, a P-value < 0.05 was used. Note that studies by the Montreal group5,6 share common participants as well as studies by the Milan group.7,8 All studies found that idiopathic RBD patients show poorer performance than healthy subjects on neuropsychological tests. However, studies are not consistent on the specific cognitive domains impaired. Several factors may explain these discrepancies. First, because most of these studies have relatively small sample sizes, the results may be affected by low statistical power. Second, there is population heterogeneity between studies on sociodemographic variables. For example, the mean educational level varied from 6 to 14 years, and the mean age varied from 64 to 70 years. Finally, the studies used different cognitive tasks, which differ in their sensitivity to detect deficits. Caution should also be taken when linking a cognitive test to a specific cognitive function, as neuropsychological tests used in clinical settings often address more than one cognitive function. For example, good visuospatial abilities as well as good planning, organization, and motor execution capacities are needed to perform well on the Rey-O figure copy test. Therefore, it would be difficult to differentiate or isolate which specific cognitive process is affected when poor performance is recorded on a given task.

Table 1. Controlled studies on cognitive performance in idiopathic rapid eye movement sleep behavior disorder
 Terzaghi et al. (2008)10Massicotte-Marquez et al. (2008)6Gagnon et al. (2009)5Marques et al. (2010)9Ferini-Strambi et al. (2004)8Fantini et al. (2011)7
  1. or share common participants. §Flexibility condition only. CTRL, controls; iRBD, idiopathic rapid eye movement sleep behavior disorder; NS, not significant; RAVLT, Rey Auditory Verbal Learning Test; SCWT, Stroop Color Word Test; TMT, Trail Making Test; WCST, Wisconsin Card Sorting Test.

Number of participants232314143240101017172412
Age67.0 ± 7.067.1 ± 6.066.6 ± 7.765.6 ± 6.565.7 ± 8.565.8 ± 8.364.0 ± 2.964.0 ± 2.070.0 ± 7.369.5 ± 7.169.5 ± 7.369.3 ± 6.3
Education6.0 ± 2.06.9 ± 2.712.2 ± 4.013.5 ± 3.913.4 ± 3.614.5 ± 2.710.0 ± 0.615.7 ± 1.78.5 ± 3.48.1 ± 3.28.6 ± 3.68.3 ± 4.0
Gender, % men91911001007853805076827575
Cognitive tasks             
Attention/executive functions
Digit span0.003NS<0.05NSNSNS
Word span<0.001
Attentive matricesNSNSNS
TMT, part B<0.03<0.05NS0.018NS
Verbal fluency      
Symbol digitNSNS
Episodic memory       
Logical memory0.0030.0220.001
 Sum of trials 1–5<0.02<0.005
 Immediate recallNS<0.03<0.005
 Delayed recall0.008NS<0.05
Grober and Buschke      
 Free recall0.012
Corsi supraspan<0.0010.025
Rey-O figure, delayed recall0.003NS
Visuospatial abilities       
Rey-O figure, copyNSNSNS<0.0010.0005
Block DesignNSNS
Bells testNS

In general, the most affected domains in idiopathic RBD are attention, executive functions, and episodic memory (characterized by impaired free recall).5–10 Additionally, some studies found impaired visuospatial and/or visuoperceptive abilities,7–9 but others did not.5,6,10 In fact, impaired visuospatial or nonverbal learning abilities in idiopathic RBD appear to be related to the extent of cognitive decline.7,11–13 Impaired visuospatial abilities have also been reported in neurodegenerative diseases associated with RBD, such as DLB and PD.5,14 Although little studied, language and praxis appear to be well preserved in idiopathic RBD.

These studies focused on group differences on cognitive tests between idiopathic RBD patients and healthy subjects. What is missing is the most relevant clinical variable – the proportion of idiopathic RBD patients who have significant cognitive impairment.


Mild cognitive impairment (MCI) is a syndrome characterized by an objectively assessed cognitive decline that is greater than expected for age and education and that does not interfere significantly with daily life activities.15 MCI refers to an intermediate state between normal cognitive functioning and dementia.15 Proposed MCI diagnostic criteria includes (i) subjective cognitive complaints reported by patients or informants at a structured interview; (ii) cognitive decline on neuropsychological assessment compared with age- and education-equivalent individuals; and (iii) preserved daily life activities based on previous and actual capacities.15,16 In addition, medication use and other medical or psychiatric conditions responsible for cognitive deficits should be excluded. MCI subtypes have been identified based on the nature (amnestic vs nonamnestic) and number (single-domain vs multiple-domain) of the cognitive domains alteration.17

In idiopathic RBD patients referred to a sleep disorders center, MCI frequency has been estimated at up to 50% (16/32) compared to 8% (3/40) in healthy individuals.5 The proportion of idiopathic RBD patients with MCI is also considerably higher than the proportion of MCI in the general elderly population, at 3–19%.15 Single-domain MCI with predominant attention and executive dysfunctions was the main MCI subtype observed in our cohort of idiopathic RBD patients.5

Individuals with MCI are at higher risk for dementia, particularly dementia associated with Alzheimer's disease, vascular dementia, and DLB.15,18 MCI progression usually follows a predictable pattern: from normal functioning to single-domain MCI and thence to multiple-domain MCI and finally dementia. Studies suggest that individuals with multiple-domain MCI are at higher risk for progressing to dementia.19,20 However, MCI progression can also be highly heterogeneous (Fig. 1). A substantial proportion of MCI patients, particularly those with single-domain MCI, return to normal cognitive functioning, whereas some remain with mild cognitive deficits for many years.15,19,21 Moreover, cognition in the elderly may be altered by many factors, including vascular diseases, respiratory disorders, psychiatric symptoms, and medication side-effects. Consequently, researchers and clinicians should be aware of the variety of progression patterns and avoid directly linking MCI to the future development of a neurodegenerative disease.

Figure 1.

Various progression patterns in patients with mild cognitive impairment (MCI).

In idiopathic RBD, the risk for patients with concomitant MCI to develop dementia has not been prospectively studied in a large cohort. Molano et al. studied seven idiopathic RBD patients with MCI who subsequently developed Lewy body disease, confirmed by autopsy.13 This suggests that idiopathic RBD patients with MCI are at higher risk for developing DLB. However, the time required to progress from idiopathic RBD to MCI and from MCI to dementia may vary greatly. For example, idiopathic RBD patients developed MCI 3–48 years after RBD onset and progressed from MCI to dementia after 1–5 years.13 Further studies are therefore needed to better understand the progression of cognitive impairment in idiopathic RBD.


A comprehensive neuropsychological assessment is the most effective way to detect cognitive impairment, including MCI. However, it is a time-consuming method that requires specialized training, and is often unavailable to idiopathic RBD patients in clinical practice. Efficient MCI screening tests in idiopathic RBD would therefore be useful. Three screening tests have been systematically assessed in idiopathic RBD:16,22 the Mini-Mental State Examination (MMSE),23 the Montreal Cognitive Assessment (MoCA),24 and the Mattis Dementia Rating Scale, second edition (DRS-2).25 The main administration features of the three tests and their psychometric properties for idiopathic RBD patients are summarized in Table 2. The MoCA and the DRS-2 show superior psychometric properties than the MMSE. However, because of its short administration time, its validated alternative versions (allowing retesting), the fact that it is available free of charge and does not require specialized training, the MoCA ( appears to be the most appropriate screening test for detecting MCI in idiopathic RBD.

Table 2. The validity of three cognitive screening tests for detecting mild cognitive impairment in idiopathic rapid eye movement sleep behavior disorder
  1. Optimal screen value: the lowest value with sensitivity and NPV at ∼80%. Optimal diagnostic value: the highest value with specificity and PPV at ∼80%. §Maximum accuracy value according to the Youden index (Y= sensitivity + specificity − 1). For PPV and NPV, estimated population base rates were 58% for MMSE and DRS-2 and 50% for MoCA. AUC, area under the curve; CI, confidence interval; DRS-2, Mattis Dementia Rating Scale, second edition; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; NPV, negative predictive value; PPV, positive predictive value.

Administration features    
Maximum score3030144
Administration time, min5–105–1015–30
Alternative versionsNoYesNo
Psychometric properties    
Cut-off scores<30<29‡§<27<26‡§<141†§<139‡§
 Sensitivity, %856088769075
 Specificity, %437962857186
 PPV, %688070838288
 NPV, %675884788371
 % correctly diagnosed686879798279
 AUC, 95% CI0.75 (0.59–0.91)0.84 (0.72–0.96)0.86 (0.73–0.98)


RBD and cognitive impairment are frequently observed features of PD.26,27 A substantial proportion of PD patients will develop dementia as the disease progresses.27 RBD in PD patients has been reported to be associated with a higher frequency of cognitive impairment.5,28,29 Indeed, we found a higher risk of having MCI in PD with concomitant RBD (73% or 16/22) compared with PD without RBD (11% or 2/18) and healthy controls (8% or 3/40).5 Moreover, we conducted a prospective follow-up study in a cohort of 42 PD patients without dementia at baseline, including 27 with polysomnographically-confirmed RBD and 15 without RBD.30 Our objective was to assess whether the presence of RBD at baseline predicted the future development of dementia. Over a mean 4-year follow-up, 48% of PD patients with RBD developed dementia, according to neuropsychological and neurological assessments. In comparison, none of PD patients without RBD developed dementia. The higher frequency of MCI and dementia in PD patients with concomitant RBD suggests that these patients are affected by a more devastating and widespread neurodegenerative disease than PD patients without RBD symptoms. However, these results remain to be confirmed by independent groups and in larger cohorts of PD patients.


Our studies have been supported by the Canadian Institutes of Health Research and the Fonds de recherche du Québec – Santé.


The authors declare no potential conflicts of interest.