Clinical profile of parkinsonism and Parkinson's disease in Lagos, Southwestern Nigeria

Parkinson's disease (PD) is a slowly progressive neurodegenerative movement disorder resulting from selective loss of nigral dopaminergic neurons of unknown aetiology. PD predominantly affects the elderly population, and is clinically characterized by both motor and non-motor manifestations. The disease occurs worldwide, with studies in Africa yielding hospital frequencies of 0.3 to 2.3% of neurological diseases [1]. In western Nigeria, the prevalence rate is between 50 - 90 per 100,000 of the population aged above ten years [2, 3]. The lower prevalence in comparison to both the black and Caucasian population in industrialized western countries such as the United States of America (USA) has been attributed to possible genetic, environmental or demographic factors [4].

The pattern of occurrence of parkinsonism (including PD) in Nigerians was comprehensively reviewed by earlier authors in Ibadan in 1979 [5], providing a basis for comparison with observed patterns globally. Their findings compared favourably with the earlier reports of Hoehn and Yahr published 12 years prior and based on cases seen in the USA [6]. Subsequently, there has been a void in reports of the pattern of PD observed in sub-Saharan African populations in recent years, particularly as the continent becomes more industrialized. The catchment area of the Lagos University Teaching Hospital, a tertiary health institution and referral center for neurological conditions, includes (but is not limited) to Lagos State in Southwestern Nigeria (a predominantly urban and industrialized area) and its environs.

This study provides a description of the clinical characteristics of our cohort of Nigerian PD patients, and provides a basis for comparison to existing observations, so as to determine if there has been any appreciable variation from previously reported patterns.

The diagnosis of the specific secondary parkinsonism was based on the constellation of clinical features suggestive of a secondary aetiology [6‐8]. Neuroimaging (brain computerised tomographic scan and magnetic resonance imaging) was available for 9 cases including the patients with olivopontocerebellar atrophy (1), hemiparkinsonism (1), 3 with vascular parkinsonism secondary to cerebrovascular disease, and 4 with PD.

A standard proforma was used to document the diagnosis of PD, personal data (present age and sex) as well as the patient characteristics (age at onset of the disease, age at presentation, duration of illness prior to presentation, and initial symptoms and distribution of symptoms of PD). The age at onset of PD-related motor symptoms was used to define age at onset of the disease in this study. Young onset PD was defined as age at onset ≤ 50 years. A positive family history of PD was defined as a history of parkinsonian features or reported physician's diagnosis of PD in a first degree relative. The severity at presentation was classified using the Hoehn and Yahr (H&Y) scale [9, 10]. PD cases presenting after December 2004 were routinely assessed in the 'off' state at presentation with the Unified Parkinson's Disease Rating Scale (UPDRS) version 3.0 (11).

Data were analyzed using EPI Info^® 2002 statistical software. Continuous data are expressed as mean (standard deviation) (SD), median, and ranges. Mean values were compared using the Student t test. The X² test was used to compare categorical variables which are presented as proportions (frequency %). P < 0.05 was considered as statistically significant.

This study provides a description of the current clinical profile of PD and secondary parkinsonism as seen over the last ten years at a tertiary referral centre in South-western Nigeria. The spectrum of cases encountered may thus be subject to referral bias, with a tendency to see more severe or more complex neurological cases. The study however provides data for comparison with an earlier report and global perspectives of parkinsonism and PD.

In contrast to the only published report on the clinical characteristics of parkinsonism in Nigerians from Osuntokun and Bademosi [5], PD accounted for a significantly higher proportion of causes of parkinsonism in the present study (79% compared to 38%). The current relative frequency of PD in our study is in keeping with available literature, which ranks PD as causing approximately three quarters of all cases of parkinsonism. The apparent disparity may be accounted for by more recent improvements and streamlining of the diagnostic criteria for PD and other parkinsonian syndromes in both the clinical and research arena in contrast to the status at the time of the earlier report [5, 12‐14].

The clinical profile of our patients differed slightly from that reported in the earlier Nigerian study. First, the mean age at onset in the present study was 61.5 years, higher than the 55.6 years reported by Osuntokun et al [5]. Both studies represent patients seen in the same geographical region, with both tertiary centres being within two hours' drive of each other. The increase in age at onset may thus be an accurate phenomenon akin to that observed by Hoehn and Yahr in which they found a time trend of increasing age at onset over decades of study [9]. It has been documented that as populations age, the age at onset of PD tends to increase. The male preponderance reported in the earlier study (ratio 4.5 to 1) was also documented here, but the magnitude in this study was lower (ratio 3.3 to 1). Although this male predilection appears to be a consistent finding, the precise reason is unknown, and the possibility of a neuroprotective effect mediated by estrogen in women exists [15‐17]. Experimental evidence indicates that estrogen may mediate this effect via several mechanisms including inhibition of dopamine transporter affinity and prevention of entry of neurotoxic agents into dopaminergic nerve terminals, thereby reducing nigrostriatal degeneration [17].

Genetic contributions to the aetiology of PD are undisputable, with transmission as an autosomal recessive or dominant trait linked to mutations in several genes, including α-synuclein, parkin, DJ1, leucine-rich repeat kinase (LRRK) 2, etc [18, 19]. Genetic susceptibility appears to be more apparent (although not exclusive) in young onset PD compared to persons developing the disease after the 6^th decade of life. We found a positive family history of PD in a first degree relative in only one of our patients who incidentally had young onset PD starting below age 50. Young onset PD was documented in 16.3% of our cohort, with a five-fold higher frequency in men (20%) compared to women (4.3%). We acknowledge the fact that the number of PD cases reported here (including only 23 females) is small and may account for the low rates of a positive family history and gender disparity in frequency of young-onset PD; our findings are thus subject to further validation. A preliminary analysis of the genetic contributions to PD in Nigerians (which included some of the cases reported in this present study), explored the role of mutations in LRRK2, PRKN and ATXN3 in apparently sporadic PD compared to age - and ethnically-matched controls. The study found that common pathogenic mutations in these genes, previously observed in several populations, are not a frequent cause of PD in Nigerians [20].

Overall, the clinical profile of PD in Nigerians does not appear to vary substantially from disease characteristics reported in other populations. Delayed presentation (and late referral) are not germaine to this study population and is one of the challenges encountered in managing PD in Africa [21]. Poor recognition of the cardinal features of parkinsonism and of the existence or benefit of available therapies in alleviating the symptoms and improving the quality of life of people with PD may contribute to late referrals. This has further implications as it will delay utilization of disease-modifying strategies which may become available in the future. Strategies to improve early recognition and referral include strengthening undergraduate movement disorders curriculum, and improving public awareness as to the existence, cardinal features, and treatment options of parkinsonism via the media. The need for such an approach is strengthened by the increased likelihood of physicians encountering PD and other neurodegenerative diseases of the elderly in the future in developing countries experiencing an epidemiologic transition marked by aging of the population.

We acknowledge the methodological limitations of this study in that follow-up data on the disease course and rate of progression, and an objective measure of magnitude of response to levodopa therapy, and mortality data which would enhance insight into the phenotypy of our PD cohort are lacking. An earlier published study of a subgroup of this cohort (28 PD cases and 28 age-matched controls) initially seen between January and June 1997 and followed up between January and May 2003 (after a 6-year interval) reported a case fatality rate of 25% in PD, compared to 7.1% in controls [22]. The peculiarity of our clinical practice scenario, characterized by the availability of a very narrow spectrum of antiparkinsonian medications and dosage formulations is reflected in the distribution of medications used in PD treatment in our cohort. We also emphasize that this factor would also limit the conclusiveness of any description of treatment outcomes in our cohort at the present time.

Our study provides data on the clinical profile of Nigerian patients with PD and demonstrates that it is similar to that from other populations, but is characterized by delayed presentation as has been reported in other developing countries. Young-onset disease occurs but may be less commonly encountered, and frequency of a positive family history is lower than in western populations.