Finger-to-Nose Test Findings in Alzheimer’s Disease

Clinical assessment of patients with cognitive impairments includes taking a history from the patient and caregiver, cognitive testing, and focused neurological examination [1]. Since Alzheimer’s disease (AD) is not associated with focal neurological deficits, the purpose of the examination is mainly to rule out other causative factors. The finger-to-nose test is routinely performed to exclude diseases associated with cerebellar involvement such as cerebellar infarction, multi-system atrophy, ataxias, etc. In this test, the patient is asked to alternately touch their nose and the examiner’s finger. The examiner’s finger is held at the extreme of the patient’s reach, and occasionally moved to a different location. Hesitation, undershooting, or overshooting on this test is termed dysmetria (or appendicular ataxia) and may suggest a lesion of the cerebellar hemispheres and associated pathways [2].

Upon returning from Fellowship with Dr. Norman Geschwind in Boston, and during his 40-year career assessing patients with cognitive impairments at the Clinique Interdisciplinaire de Mémoire (CIME) du CHU de Québec, neurologist Dr. Rémi W. Bouchard has come to notice that patients with AD, without exhibiting the classic dysmetria associated with cerebellar disorders, showed a strikingly prototypical impairment on the finger-to-nose test. First, he noticed that the task was in itself an excellent test for limb apraxia, with many patients not able to understand or execute the task. Interestingly, Dr. Bouchard noticed that, once they succeed in understanding and performing the task, patients with AD are often at a loss when asked to repeat the task with the contralateral limb, a finding he termed the ‘second finger syndrome’. He also noted that AD patients tend to abnormally push on the examiner’s finger before returning back to their nose— sometimes in a repetitive manner— a phenomenon he termed the ‘distal pressure sign’. This sign is distinct from dysmetric overshooting seen in cerebellar disorders in that there is no hesitation before reaching the target.

We retrospectively reviewed records of patients seen between January 2013 and July 2016 at CIME to estimate the prevalence of such findings in various neurocognitive syndromes. We looked specifically at three diagnostic categories: AD (n = 365), vascular cognitive impairment (VCI; n = 40), and subjective cognitive impairment (SCI; n = 56). All diagnoses were performed by experienced behavioral neurologists based on state-of-the-art clinical work-up including clinical evaluation, neuropsychology, magnetic resonance imaging, and in some cases ¹⁸F-fluorodeoxyglucose positron emission tomography (FDG-PET) as well as cerebrospinal fluid analyses for AD biomarkers or amyloid PET. Close to 80% of patients’ records contained a rating of the finger-to-nose test. We found that 74% of AD patients showed cognitive abnormality on the finger-to-nose test, compared to 14% in VCI and 2% in SCI (see Table 1). Conversely, patients with VCI had more cerebellar dysmetria (10%) than those with AD (<1%). Patients with a cognitive abnormality on the finger-to-nose test had a significantly lower Mini-Mental State Examination (MMSE) score (20.7±5.8) than those without (25.3±4.2; p < 0.01), suggesting that it might be more a marker of severity than etiology. Nevertheless, cognitive abnormality on the finger-to-nose test remained more prevalent in AD than VCI regardless of the severity, even in the earliest stages (69% in AD versus 6% in VCI in patients with MMSE >25). Interestingly, 8/13 (61%) patients with the logopenic variant of primary progressive aphasia (also called the language variant of AD) showed abnormality on the finger-to-nose test, compared to 3/11 (27%) patients with disorders along the frontotemporal lobar degeneration (FTLD) spectrum (behavioral variant FTLD [n = 7], progressive non-fluent aphasia [n = 2], semantic dementia [n = 2]). Among the AD group, the most prevalent abnormalities were the second finger syndrome (137/289 [47%]) and the distal pressure sign (131/289 [45%). In VCI and FTD, distal pressure was the most prevalent (6/40 [15%]).

Clinicoanatomical localization of these signs is still unclear. While recent authors have highlighted the potentially overlooked role of the cerebellum in AD symptomatology [3, 4], we believe that cognitive finger-to-nose test abnormalities do not result from cerebellar dysfunction. It is known that patients with AD are often unable to imitate meaningless gestures and pantomime the use of tools— a clinical feature called limb apraxia [5, 6]. The inability to perform the finger-to-nose test with the first or second hand may represent a form of limb apraxia, resulting from the disruption of the dorsal praxis streams involving superior and medial parietal cortices [6, 7]. Meanwhile, the distal pressure sign might be a form of perseverative behavior— analogue to the well-known applause sign [8]— resulting from disruption of frontostriatal networks subserving behavioral inhibition and executive functions [9]. More research in needed to clarify the anatomical substrates of the second finger syndrome and distal pressure sign, as well as their relation to known clinical features such as limb apraxia and the applause sign.

These findings are limited by the fact that the only clinician who systematically reported finger-to-nose findings in his patients’ charts is Dr. Rémi W. Bouchard. Further studies are needed to assess the inter-rater reliability of this observation. We must also emphasize that such signs are not meant to have strong sensitivity or specificity for AD. Similar to the head-turning sign, also described by Dr. Bouchard [10] and whereby AD patients tend to turn to their caregiver when asked questions, the ‘second-finger’ and ‘distal pressure’ signs add to the clinician’s toolbox when assessing patients with cognitive disorders without adding time to the consultation. Future research should examine the inter-rater reliability of these observations as well as their diagnostic properties in prospective cohorts.