Parkinson’s disease (PD) is a chronic debilitating disease, complicated by cognitive dysfunction and associated with increased caregiver burden, pressure on community health facilities and significant mortality (Okubadejo et al., 2006; Akinyemi et al., 2008). Most of the data concerning cognitive dysfunction in PD are based on studies done in Europe and North America. There is paucity of data on cognitive impairments among PD patients in Sub-Saharan Africa. Knowledge of the pattern of cognitive dysfunction and predictive factors perculiar to different geographical regions may be important in early identification of this complication and ensuring appropriate channeling of resources into caring for patients with PD.

Cognition refers to all mental activities associated with thinking, knowing, remembering and communicating. It embodies solving problems, making decisions, mental grouping of objects and forming judgments (Lezak et al., 1995). The major classes of cognition include reception (ability to select, acquire, classify and integrate information), memory and learning (which involve information storage and retrieval), thinking (involving mental organization and recognizing the information), and expressive functions (which involve speaking, drawing, writing, physical gestures, facial expressions and movements) (Lezak et al., 1995).

The prevalence of cognitive impairments in PD has been reported in several developed countries. A systematic review of 27 studies by Cummings reported an average prevalence of 40% while Aarsland et al. (2007) in their systematic review found the prevalence of cognitive impairments in PD to range between 24 and 31% (Cummings, 1988a; Aarsland and Kurz, 2010). The UK CamPaIGN study obtained a prevalence of 36%, while Muslimovic et al. (2005) found a prevalence of 24% in a cohort of newly diagnosed PD patients in the Netherlands (Foltynie et al., 2004). One of the few studies from the sub-Saharan Africa by Osuntokun and Bademosi (1979) found a frequency of 5% but a more recent study by Akinyemi et al. (2008) reported a frequency of 21.6%.

Several studies, mostly from developed countries, have described cognitive domains affected in PD. Such domains include memory, language functions, attention and concentration, mental speed, motor fluency and praxis. As regards the predictors of cognitive dysfunction in PD patients, most studies agreed on the role of disease severity and presence of dyspraxia. There are conflicting reports on the predictive role of age at onset, duration of disease and level of education of the affected patients on cognitive impairment. It has also been argued that since PD is common among the elderly people (>65 years), the age-related cognitive impairment may be contributing to the cognitive dysfunction in PD thereby confounding the significance of age at onset of the disease. In the light of this, there is need to evaluate the disease and patient variables that can predict development of impaired cognition in PD.

The dearth of information on the pattern and predictive factors of cognitive dysfunction in sub-Saharan African region prompted this study.

Therefore, this study sought to determine the frequency and pattern of cognitive dysfunction amongst patients with PD presenting at a tertiary health facility in south-south Nigeria and compared the performance of the PD patients with demographically matched subjects without PD (controls). In addition, we assessed the contributory roles of patient- and disease- specific variables to the development of cognitive impairments in PD patients.

This case control study compared the cognitive performances of 30 PD patients with age, sex, and level of education matched controls using the CSID and demonstrated that cognitive impairments are more common in PD, also showed the predictive significance of apraxia, severity of disease based on Hoehn and Yahr staging and UPDRS score, and presence of recall deficiency, and poor naming ability.

The patients were in the sixth decade of life. This may be due to the fact that PD is rare before the age of 50 years with a sharp increase in incidence after the age of 60 years (Ericksen et al., 2005; De Lau and Breteler, 2006). Majority of the patients that presented during this period were males and this is consistent with earlier studies that have reported a higher prevalence of PD in men than in women (Li et al., 1985; Mayeux et al., 1995; Fall et al., 1996; Claveria et al., 2002; Benito-Leon et al., 2003; Okubadejo et al., 2010). It was also observed that most of the PD patients presented in stages 2 and 3 of the Hoehn and Yahr staging, this may be due to the fact that most patients with PD do not present early; probably due to lack of impairment in activities of daily living in the early stage of the disease. The fewer number in the later stages may be because severe cases might have died either from complications of the disease or other age-related illnesses.

The PD cohort studied had significant cognitive impairment as demonstrated in their performances on all the neuropsychological tests used. This is similar to findings in previous studies on cognitive impairment in PD. The frequency of cognitive dysfunction in the PD patients was 50% using the CSID. This is higher than those obtained in other studies (Foltynie et al., 2004; Akinyemi et al., 2008; Aarsland and Kurz, 2010). The variation in prevalence of PD from this study may be attributed to differences in methods of cognitive assessment, operative definition and the study population (Emre, 2003a). This study was however, on cognitive impairment and not just dementia alone.

The CSID assessed cognitive domains of attention and calculation, orientation to time and place, memory, language, executive function, psychomotor speed and constructional apraxia. The overall performance of patients with PD on the test instrument was significantly worse when compared with the performance othe healthy subjects. The CSID demonstrated cognitive dysfunction in the domains of memory, language, executive function and visuospatial dexterity. This is in keeping with earlier studies done (Hall et al., 1993, 1999; Imarhiagbe et al., 2005). The patients overall performance in the memory cognitive domain was poor compared with the controls. Memory deficit in PD is more a problem of retrieval of coded information than a deficit of registration, encoding and storage of information (Tomer and Rey, 1992; Okubadejo et al., 2006; Salmon and Filoteo, 2007). This was demonstrated in this study by the performance of the PD subjects in the name recall, short story test, past national events and overall memory score. Their performance was worse than that of the controls.

Language deficits have been well-described in PD and these include naming difficulties, decreased information content of spontaneous speech, impaired comprehension of complex sentences, as well as verbal fluency (Bayles et al., 1993; Péran et al., 2003). These were exhibited in our cohort of subjects. Confrontational naming defects exhibited by our subjects may be as a result of direct neuronal degeneration associated with hippocampal and limbic deposition of Lewy body pathology in PD (Braak et al., 2003). However, in many described language deficits, the anatomical regions of the brain responsible for language function are not primarily affected by the disease process, rather they may be related to the dysequilibrium syndrome that characterizes PD (Starkstein et al., 1989; Emre, 2003b). In addition, patients with PD may develop a limbic or hippocampal type of amnesia similar to that seen in Alzheimer’s disease. Such subjects are those in whom AD and PD co-exist or those with AD type pathology (Cummings, 1988b).

The patients with PD performed worse than the controls in the tests assessing executive dysfunction such as the animal naming fluency test (1 min) and comprehension-motor response. The timed test offers some quantitative measure of the speed of cognitive processing (reduced in PD). PD patients are impaired in tasks requiring the spontaneous elaboration, maintenance and change of cognitive strategies (Bowen et al., 1975; Matison et al., 1982; Weingartner et al., 1984; Flowers and Robertson, 1985; Taylor et al., 1986; Gotham et al., 1988). The short story test as a test of working memory (a recognized subcategory of executive function) (Baddeley, 1992; D’Esposito et al., 1995) was impaired in the PD subjects.

According to Farina et al. (1994) in PD the dysfunction of the “Complex Loop,” linking pre-frontal lobe to the basal ganglia, is combined with damage to the “Motor Loop,” a functional anatomic circuit connecting the putamen to the supplementary motor area, resulting in disruption in functions of the ascending dopaminergic mesocorticolimbic pathway and frontal cortex responsible for the motor disability and cognitive symptoms. The frontal-like neuropsychological abnormalities observed in PD in another hypothesis was attributed to the loss of dopaminergic neurons in the ventral tegmental area; affecting meso-cortico-limbic pathways, causing dopamine depletion in various cortical areas, including the pre-frontal cortex (Lees and Smith, 1983; Torak and Morris, 1986; Sharpe, 1990).

In this study, certain variables were demonstrated to predict cognitive dysfunction. Although the age groups did not significantly correlate with the CSID, and thus was not related to the level of overall cognitive function. The Campaign cohort study revealed that age greater than 72 years was an independent predictor of cognitive decline in PD patients (Williams-Gray et al., 2009a).

Age at onset of the disease and disease duration were not found to be predictors of cognitive dysfunction. This is in keeping with Hughes et al. (2000) who found no relationship between age at onset of the disease and cognitive decline. Aarsland et al. (2007) however, showed that it is the general effect of age rather than the age of PD onset that contributes most to the incident dementia in patients with PD, while Lyros et al. (2008) showed that the cognitive decline in PD patients might be owing to simultaneous effect of age-related and disease associated neuropathology. This was at variance with Akinyemi et al. (2008) who demonstrated that older age at onset of PD showed significant association with cognitive dysfunction.

Previous studies have shown that the most important clinical predictors of global cognitive decline following correction for age were neuropsychological tasks with a more posterior cortical basis including semantic fluency less than 20 words in 90 s and inability to copy an intersecting pentagon figure, as well as a postural instability gait abnormality (non-tremor dominant) motor phenotype at the baseline assessment (Katzen et al., 1998; Levy et al., 2002; Azuma et al., 2003; Williams-Gray et al., 2009a,b; Aarsland and Kurz, 2010; Taylor et al., 2008). This study also demonstrated that semantic fluency and inability to copy an intersecting pentagon were significant predictors of cognitive dysfunction, in addition recall was also significant, however, was at variance with the previous studies in that gait abnormality was not found to be a significant predictor of cognitive dysfunction.

We demonstrated that greater severity of PD as indicated by higher Hoehn and Yahr stages or higher scores on the UPDRS motor subscales was associated with worse cognitive performance which is in tandem with previous studies (Blessed et al., 1968; Muslimovic et al., 2005; Akinyemi et al., 2008).

There was no significant relationship between years of formal education and cognitive decline. This is in keeping with studies by Akinyemi et al. (2008) and Pai and Chan (2001). However, Glatt et al. (1996) and Green et al. (2002) found an association between lower educational attainment and cognitive decline in subjects with PD.

It is important to mention that there are a few studies that used Mini-Mental State Examination (MMSE) to assess cognitive impairment among Nigerian patients with PD (Ojo et al., 2012; Ojagbemi, 2013), but the CSI-D is more robust in testing for the various cognitive domains and allow for a wider comparison of findings considering its wider acceptance among researchers.

The limitations of this study included potential recall bias and inability to calculate incidence, which are common to case-control studies. However, considering that PD is relatively uncommon in our environment, case control study design was deemed suitable for this research. The sample size is an obvious limitation of our study. Although we calculated the sample size statistically, a larger sample size will provide more generalizable conclusions.

This study showed that there is a high frequency of cognitive impairment amongst patients with PD compared to demographically matched controls and this worsens with the stage of the disease on Hoehn and Yahr staging and with increasing UPDRS score. Other predictors of cognitive impairment in the PD patients included semantic fluency, apraxia and recall. The pattern of cognitive impairments seen in these patients included memory, psychomotor speed, complex mental speed and visuoperceptual and constructional dexterities.

All PD patients should be screened for cognitive impairment using neuropsychological tests by their health care providers.

Most of these PD patients are dependent on others for means of livelihood due to incapacitation by the disease. There is need for support groups such as the Parkinson’s society in our local environments.

Further studies are recommended to increase knowledge of this disease and the impact on the cognitive function of affected individuals. This will help in the management of these individuals.

The original contributions presented in this study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by University of Benin Teaching Hospital Research Ethics Committee, Benin City, Nigeria. The patients/participants provided their written informed consent to participate in this study.

EO and OO conceptualized, designed, and approved the manuscript. FA analyzed the data and approved the manuscript.