Abstract

Importance: Olfactory dysfunction has been well-described in Alzheimer’s disease (AD). Options for improving olfaction are limited, and few studies have prospectively attempted to improve hyposmia. A previous meta-analysis has shown that exercise provides global benefit to patients with AD.

Objective: We aim to prospectively evaluate what effect exercise has on olfaction and general cognition in patients with AD.

Design: Sub-study of a randomized, controlled clinical trial of 26 weeks duration.

Setting: Tertiary referral center.

Participants: Individuals with early-stage AD and adequate visual and auditory abilities to perform all cognitive testing were included into this study. Only individuals with MMSE of 16-30 and scored 0.5 (very mild) or 1.0 (mild) on the Clinical Dementia Rating Scale (CDR) were enrolled.

Interventions: Participants were randomized into either an aerobic exercise program or stretching program. Aerobic exercise typically consisted of treadmill walking for 150 minutes per week with similar times of stretching exercises for the control arm.

Main Outcomes and Measures: Patients completed the UPSIT (University of Pennsylvania Smell Inventory Test) and MMSE (Mini-Mental State Examination) prior to and after the 26 weeks program.

Results: This study included 18 patients with 9 patients in the control arm and 9 in the exercise arm. The groups showed no statistical difference (p=0.537) between pre-treatment and post-treatment olfactory scores. There were 3 patients in the exercise arm versus 1 patient in the control arm that had an improvement in olfactory score (p=0.288). The groups showed no significant difference in post-intervention average MMSE (p=0.884), though more patients improved in the exercise arm.

Conclusion: Although more patients in the exercise arm versus the control arm improved in normative olfactory score and MMSE scores, it may be hard to predict statistical significance due to small sample size. This is suggestive that exercise may play a role in aiding certain patients. Reproduction of this study with a larger sample size and shorter intervention length may provide further clarity.