Differential diagnosis of unexplained falls in dementia: Results of "Syncope & Dementia" registry

Background: Dementia patients have an increased risk of fall, and some of them might suffer from undiagnosed syncope. The present analysis aimed at identifying predictors of differential diagnosis between syncopal and non-syncopal fall in patients with dementia included in the “Syncope & Dementia” registry. Methods: We enrolled patients aged 65+ with a diagnosis of dementia and a history of syncope and/or unexplained fall. All subjects underwent a comprehensive geriatric assessment, including the syncope protocol of the European Society of Cardiology. Subjects whose syncope diagnosis was confirmed were labeled as “Confirmed Syncope” (CS). Patients with unexplained fall were labeled as “Syncopal Fall” (SF), if a final diagnosis of syncope was performed, or as “Non-Syncopal Fall” (NSF), if syncope was excluded. Results: We included 372 subjects (mean age 84, 61% females). Mini Mental State Examination score was higher among SF (18.5 ± 4.9) compared to NSF patients (15.6 ± 5.8, p= 0.02). In a multinomial logistic regression model with NSF as the reference group, CS patients less often suffered injuries and more often reported history of syncope, while patients with SF had a better cognitive status and were more often exposed to precipitating factors, including postural changes and neck movements. The absence of prodromes and the intake of benzodiazepines and insulin was highest in NSF patients. A simple score including main clinical predictors showed an 82% sensitivity with a 56% specificity in discriminating SF from NSF patients. Conclusion: Simple clinical markers can aid in the differential diagnosis of unexplained falls in dementia, separating syncopal from non-syncopal falls.