Establish a Nomogram to Predict Falls in Spinocerebellar Ataxia Type 3

Abstract

Purpose: Falls are common and are frequently accompanied by injuries in patients with spinocerebellar ataxias type 3 (SCA3). We explored which factors could predict falls in a cohort of patients with SCA3 and developed a nomogram model to predict the first fall in non-fallen patients with SCA3.Method: We conducted a prospective cohort study. Forty-four non-fallen patients with SCA3 were followed up until the first fall or November 5, 2020, whichever came first. Univariate and multivariate Cox proportional hazard regression analyses were applied to explore the predictive factors of falls in patients with SCA3. A nomogram model predicting the no-fall probabilities at 3, 6, 12, and 24 months was formulated based on the results of the multivariate Cox analysis. Internal validation was conducted to assess the discrimination and calibration of the final model using bootstrapping with 1,000 resamples.Results: Multivariate Cox proportional hazard regression showed that the presence of dystonia, hyperreflexia, urinary incontinence, and hidrosis and the number of abnormal eye movements predicted a more rapid progression to falls in patients with SCA3. The nomogram model showed good discrimination with a concordance index of 0.83 and good calibration.Conclusion: Patients with dystonia, hyperreflexia, urinary incontinence, and hidrosis, and more types of abnormal eye movement had a more rapid progression to falls in SCA3.