Path Integration Detects Prodromal Alzheimer’s Disease and Predicts Cognitive Decline

Abstract

Background: The entorhinal cortex is the very earliest involvement of Alzheimer’s disease (AD). Grid cells in the medial entorhinal cortex form part of the spatial navigation system. Objective: We aimed to determine whether path integration performance can be used to detect patients with mild cognitive impairment (MCI) at high risk of developing AD, and whether it can predict cognitive decline. Methods: Path integration performance was assessed in 71 patients with early MCI (EMCI) and late MCI (LMCI) using a recently developed 3D virtual reality navigation task. Patients with LMCI were further divided into those displaying characteristic brain imaging features of AD, including medial temporal lobe atrophy on magnetic resonance imaging and posterior hypoperfusion on single-photon emission tomography (LMCI+), and those not displaying such features (LMCI–). Results: Path integration performance was significantly lower in patients with LMCI+than in those with EMCI and LMCI–. A significantly lower performance was observed in patients who showed progression of MCI during 12 months, than in those with stable MCI. Path integration performance distinguished patients with progressive MCI from those with stable MCI, with a high classification accuracy (a sensitivity of 0.88 and a specificity of 0.70). Conclusions: Our results suggest that the 3D virtual reality navigation task detects prodromal AD patients and predicts cognitive decline after 12 months. Our navigation task, which is simple, short (12–15 minutes), noninvasive, and inexpensive, may be a screening tool for therapeutic choice of disease-modifiers in individuals with prodromal AD.