Positional vertigo in a falls service

SIR—Dizziness is one of the commonest symptoms described by older people [1] and is associated with balance disorders, functional decline, reduction in quality of life and falls [2–4]. At least 50% of older people with dizziness complain of two different dizzy symptoms; the most common being a gait disorder (a feeling of disequilibrium on walking) [5], and the other arising from the cardiovascular and peripheral vestibular systems [6]. Dizziness and balance disorders in older people are common presenting symptoms at Falls clinics. ‘Positional vertigo’ (brief episodes of vertigo with nystagmus provoked by changes in head position) is one of the most common symptoms [7] presented to an ear, nose and throat (ENT) service. Benign paroxysmal positional vertigo (BPPV) accounts for the great majority of these cases and is caused by peripheral vestibular rather than central pathology [8]. Posterior canal benign paroxysmal positional vertigo (p-BPPV) is the commonest type of positional vertigo seen and is treatable [9]. Prevalence of BPPV increases with age [7] and in one study on medical outpatients with unrecognised BPPV, it has been found to be associated with a history of falls [10]. Older adults with BPPV are more likely to present to falls services managed by geriatricians rather than an ENT department as they frequently describe postural ‘dizziness’ rather than vertigo and have multiple types of dizziness [11]. The aim of this study is to identify (i) patients presented to a Falls and Syncope Service (FSS) with positional vertigo and nystagmus and (ii) compare and contrast the prevalence of falls within that group to those presenting with other dizzy symptoms to a Falls and Syncope Unit. The FSS based in Newcastle serves over 1,500 new patients annually, with one third describing dizziness alone or dizziness in combination with symptoms of falls and/or syncope. There is a separate, regional ENT service in Newcastle offering diagnostic and treatment service for vertigo and is available for all ages. We identified consecutive patients seen in FSS by a single observer over a 2-year period between January 2005 and 2007. These patients were referred to FSS with dizziness as the main presenting complaint. Types of dizziness included presyncope, unsteadiness, light-headedness and non-specific dizziness. Standardised assessment was performed in each individual and is described in Figure 1. Gait disorders were categorised using the Nutt classification [12]. Schematic representation of the investigation and management of patients attending our falls and syncope clinic with dizziness Diagnosis of BPPV was made when typical signs (nystagmus) and symptoms (vertigo) are provoked by positional tests such as the Dix-Hallpike test or side-lying test [13]. The diagnosis of p-BPPV, the most common type of BPPV, was made when there was positional nystagmus with torsional and vertical components appearing when the patient was positioned to the symptomatic side [7]. Positional vertigo arising from central pathology is less common and is suspected when clinical features differed from BPPV, e.g. latency, duration and particularly the direction of nystagmus, and if additional neurological signs were present [14]. Those with positional vertigo along with nystagmus (PN group) from either BPPV or central positional nystagmus (CPN) were compared to a group of consecutive dizzy patients assessed using the same standardised criteria by one observer but found not to have PN (non-PN group). Comparisons between groups were done using parametric tests for continuous variables, unpaired t-tests and Fishers exact test for proportions. A statistically significant result was when P<0.05. Over a 2-year period, 850 patients complaining of dizziness were included. Median duration of dizzy symptoms was 12 months (range 3–144). A total of 123 (14.5%) subjects had positional nystagmus (PN) on Dix–Hallpike or side-lying test (the PN group). Mean (range) age of the PN group was 71 years (36–94) with 86 (70%) females and were comparable to the non-PN group, with mean (range) age of 64 years (26–91) with 72% females (P = ns). In 103 (84%) patients, the positional test used for diagnosis was the Dix-Hallpike test. This test was not suitable (because of limitations in cervical spine movement) in the remaining 20 (16%) where the side-lying test was used. One hundred and ten subjects (89%) of the PN group had nystagmus characteristic of p-BPPV. In 63 (51%) the diagnosis was made for right p-BPPV and in 40 (32%) for left p-BPPV, with 7 (6%) having bilateral p-BPPV. Thirteen (11%) subjects had nystagmus on positional testing which was asymptomatic and characteristic of a central vestibular cause. There were no differences between these phenotypes in age, duration of symptoms or presence of additional gait abnormality or cardiovascular type of dizziness. Fifty-three (44%) subjects in the PN group had fallen compared to 28% of the non-PN group (P = 0.04). Fallers were significantly older than non-fallers (P = 0.01). When the entire cohort of PN subjects was separated into the younger group (aged 30–49), the middle aged (aged 50–59) and older group (older than 70 years), no differences were found in the proportion who were males, duration of symptoms, the position in which nystagmus was induced, prevalence of gait abnormality and concomitant cardiovascular diagnosis. Falls were, however, significantly more frequently reported in the older age groups (P = 0.001) (Table 1) Comparison of falls in those with positional vertigo with nystagmus between different age cohorts and coexisting types of dizzy symptoms Comparison of falls in those with positional vertigo with nystagmus between different age cohorts and coexisting types of dizzy symptoms Although fallers in our PN cohort were not significantly more likely to have a gait abnormality, there was a trend for it to be more prevalent with increasing age. Details of the injuries sustained by fallers are presented in Appendix 2 in the supplementary data on the journal website (http://www.ageing.oxfordjournals.org). Thirty (24%) of 123 patients in the PN group had an additional cardiovascular type of dizziness. Routine cardiovascular investigations confirmed Vasovagal syncope in 10, orthostatic hypotension in 6, both in 2, bradyarrhythmias in 3 and probable cardiac arrhythmias in 3 with implantable loop recorders now in situ. (Table 1) Fifty-nine (48%) patients complained of disequilibrium and had a gait disorder in addition to PN. This was felt to be of lower level gait type in 33, of middle level type in 24 [cerebellar (4) and strokes (5)] and 2 with a high-level gait disorder. All subjects with a significant gait disorder were offered balance rehabilitation. All 110 patients with p-BPPV were offered Epley manoeuvre but this was difficult in 32 (29%) because of co-morbidity including arthritis of cervical spine, shoulder problems and significant frailty. In those with central nystagmus, no diagnosis has ultimately been made to account for the presence of positional nystagmus in 5 (39%), and other diagnoses are shown in Appendix 3 in the supplementary data on the journal website (http://www.ageing.oxfordjournals.org). A study from medical out-patients [10] suggested that individuals with undiagnosed BPPV reported more falls than those without BPPV. The current study, performed in a falls service, confirms that those with positional nystagmus are more likely to fall than those with other types of dizziness and the oldest patients with PN are more likely to sustain significant injuries than younger patients. In our study, 38% of those with PN fell as a direct consequence of BPPV describing falls precipitated by looking up, bending over or with postural change. Half of those with BPPV in community studies complain of a sensation of imbalance seemingly unrelated to any precipitating head movements [15] and show impairments in balance testing [16], a difference more pronounced in older patients [17]. Our results confirm previous studies that p-BPPV is the commonest cause of positional nystagmus [8]. Horizontal and anterior canal BPPV are rarer; we saw neither of these subtypes in this large series presented to a falls service. In this series it is important to note that patients with CPN, although less common (11% of the total group), can be presented to a falls unit rather than directly to neurology services. This prevalence is similar to the 12% seen in the series from an otolaryngology clinic [18]. Sustained, downbeat nystagmus appears to be the most frequent type of CPN seen [14]. This was the commonest diagnosis made in our central group and was associated with cerebellar pathology. This underlines the importance of ensuring appropriate training of clinicians working in falls units who must have skills to differentiate types of nystagmus. BPPV leads to significant morbidity, psychological impact and medical costs in all age groups [15] and health-related quality of life in elderly patients [19]. Dix-Hallpike test and side-lying test are validated as cost-effective tests for the diagnosis of BPPV [13, 20]. There were no complications in performing these tests in more than 500 older patients in a falls unit. This study shows that half of all the patients presented with positional nystagmus to a falls unit also had other types of dizziness requiring specific investigation and management. This suggests that older patients with dizziness may require a comprehensive service perhaps based in a falls unit, which should include expertise in diagnosing and treating BPPV as well as access to cardiovascular testing and accurate diagnosis and rehabilitation of gait disorders. Dizziness in older persons is a common presentation to a falls service. Older persons with BPPV are more likely to fall than with any other type of dizziness. BPPV is a treatable risk factor for falls. Older persons have multiple types of dizziness requiring a structured, comprehensive assessment which should include positional testing for BPPV. There was no conflict of interest or funding given to support this research.