EP 99. Prediction of falls through gait variability and symptom severity in cerebellar syndromes

Abstract

Falling is a major problem in patients with cerebellar ataxia. Recently, increased gait variability was found to be associated with a history of falls. However, due to the complex and multifactorial nature of falling, falls prediction and prevention remains a challenging task for healthcare givers. The aim was to first investigate and categorize the circumstances of falls, and to then examine if increased gait variability and symptom severity can predict the occurrence of different types of falls in patients with cerebellar syndromes. This work is part of a prospective cohort study on fall risk in patients with vertigo and balance disorders (trial number: DRKS00006243). At baseline, a complete clinical examination is conducted, including scoring of disease severity (SARA score) and assessing gait and balance. Gait is tested using a GAITRite® sensor mat. During the following year, patients keep a fall diary and receive monthly telephone interviews about their fall events. Preliminary data of 24 patients with cerebellar syndromes during a 6-months observational period is presented. Mean age and SARA score were 47.5 ± 14.0 and 12.2 ± 4.4, respectively. Twenty out of 24 patients reported on a total number of 135 falls (including near-falls). The most common three fall categories were falls due to (1) transferring from one position to another (n = 22 of 135 falls), (2) tripping while unperturbed walking (n = 21 of 135 falls), and 3) managing stairs (n = 20 of 135 falls). Only falls due to tripping while unperturbed walking were associated with an increase in stride time CV (OR: 3.9 [1.2; 12.5]) and stride length CV (OR: 2.5 [1.1; 6.1]). Gait variability was positively correlated with the SARA score (Table 1). Accordingly, higher SARA scores were only predictive for falls due to tripping while unperturbed walking (OR: 2.0 [1.1; 4.0]). Two risk factors have been identified to be predictive for the occurrence of falls due to trips while unperturbed walking. Other types of falls, such as falling on stairs and falling while performing transfers, could not be predicted by the same factors. Identification of specific risk factors for different types of falls is necessary to develop target-oriented fall prevention measures.