Introduction: Cervical dystonia (CD) is the most common form of adult-onset focal dystonia. Because of a heterogeneous clinical presentation, the diagnosis rests on clinical opinion. During the last decades, several clinical trials have tested safety and efficacy of medical and surgical treatments for CD. We analyzed all the published CD trials and reviewed the strategies adopted for patient enrollment.Methods: The review included clinical trials in patients with CD published in PubMed. Studies were excluded if reviews, meta-analyses, post-hoc analyses on pooled data, or if not reporting a treatment for CD.Results: A total of 174 articles were identified; 134 studies met inclusion criteria. Diagnosis of CD varied among studies and in most cases was based on clinical judgement, using different descriptors such as "cervical dystonia " (37 studies), "idiopathic or isolated CD " (35), "primary CD " (13), and "torticollis " (40). Clinical judgement was supported by a phenomenological description of dystonia in four studies, and by a specific diagnostic strategy in other four. Finally, one study adopted general diagnostic criteria for dystonia. Inclusion and exclusion criteria proved heterogeneous across trials and were defined only in 108 studies, mainly considering age or the phenomenological pattern of muscle involvement.Conclusion: The review showed lack of consolidated diagnostic criteria and non-uniformity of eligibility criteria for CD across clinical trials. There is need to move beyond clinical judgement as diagnostic criterion for selecting participants. New trials assessing specific CD patient subgroups or comparing medical and surgical procedures will need grounds that are more consistent.