Background and purpose: Few studies have attempted to develop clinical predictors for cervical dystonia (CD) aiming at progression of the dystonic movement.
Method: We retrospectively evaluated 73 patients with primary CD who underwent treatment with Botulinum toxin type-A (BTX-A). The patients were assembled in two groups according to the spread of dystonia during follow-up: spreading and non-spreading CD. We performed a binary logistic regression model using spreading of cervical dystonia as dependent variable aiming to find covariates which increase the risk of spreading.
Results: Our logistic regression model found the following covariates and their respective risk ratios: time of disease >18.5 months=2.4, retrocollis=1.9, anterocollis=1.8, head tremor=1.6.
Conclusion: Time of disease >18.5 months, retrocollis, anterocollis and head tremor may predict spreading of dystonic movement to other regions of the body in CD patients.