We included 23 patients with ACD who frequented the Botulinum Toxin and Movement Disorders Outpatient Unit of the Fondazione Policlinico Universitario A. Gemelli IRCCS and collected clinical assessment and data about their response to the BoNT/A treatment.

Inclusion criteria were: (1) a diagnosis of ACD dystonia; (2) at least six consecutive treatments with one of the commercially available types of BoNT/A injections, i.e., onabotulinumtoxinA (A/Ona, Botox®, Allergan Inc, Irvine, CA, USA); abobotulinumtoxinA (A/Abo, Dysport®; Ipsen, Slough, Berkshire, UK); incobotulinumtoxinA (A/Inco, Xeomin ®; Merz Pharma, Frankfurt, Germany); (3) age ≥ 18 years.

Exclusion criteria were: (1) diagnosis of idiopathic or genetically proven CD; (2) inability to retrieve data; (3) previous surgical treatment for dystonia; and (4) failure to sign the informed consent form.

In all patients, ACD diagnosis was confirmed by the coordinator of the Movement Disorders Unit. The diagnosis of ACD followed the etiological axis [2]: drug-induced; associated with Parkinson’s Disease (PD), other parkinsonism or other movement disorders; post-traumatic; metabolic; associated with CP, brain tumors or other neurological conditions.

Patients were classified according to their clinical CD presentation (torticollis, laterocollis, retrocollis, anterocollis or combined forms, dystonic tremor, shoulder elevation) and whether they had focal, multifocal, or generalized dystonia or hemidystonia. The procedures were in line with the ethical standards of the Helsinki Declaration (1964, amended most recently in 2008) of the World Medical Association.

This study evaluated the long-term use of flexible dosing regimens of botulinum toxin in a setting close to real-life clinical practice where the dose used is determined based on clinical needs. So, our Patients with ACD received injections of botulinum toxin A using flexible intervals (3–6 months) and dosing based on their needs and not in a fixed pattern in terms of time-lapse and units injected. The treatment was adjusted according to the therapeutic results and AE at each visit. BoNT/A was reconstituted in sterile saline solution sodium chloride as follows: (1) 100 U A/Inco in 2 cc (A/IncoBoNT concentration of 5 U per 0.1 cc); (2) 100 U A/Ona in 2 cc (A/OnaBoNT concentration of 5 U per 0.1 cc; (3) 500U A/Abo in 2.5 cc (A/AboBoNT concentration of 20 U per 0.1 cc). As in ICD, the starting doses were chosen according to the severity of dystonia, keeping the first dose as low as possible to minimize the risk of side effects.

In line with standardized procedures, a neurologist who was highly trained in movement disorders and BoNT use performed most of the treatments with instrumental targeting such as electromyography (EMG) guidance to increase the accuracy of the BoNT injections [15] and to detect activity and target posterior and deep neck muscles, with an expected amelioration of outcome [16]. The following muscles were injected based on the prominent dystonic pattern: trapezius, sternocleidomastoid (SCM), scalene, splenius capitis and levator scapulae.

We collected the following data: (1) treatment parameters (number of BoNT/A treatments; type of BoNT/A toxin; total dose and single muscle doses; EMG or US-guided targeting); (2) efficacy of the treatment in terms of latency, total duration, and duration of peak effect at each follow-up visit as reported by patients and/or next of kin. Latency was defined as the interval between injection and the first sign of improvement noticed by the patient. The total duration of improvement was defined as the interval between the first day of improvement and the last day of reported benefit; the peak effect duration was the number of days the patients experienced the best clinical effect. (3) In keeping with our previous studies [8, 17], therapeutic response as a global clinical assessment (GCA) was based on patient’s and/or of next of kin’s perception of improvement expressed as 0–100% compared with the baseline condition. CGA was rated from 0 to 6 (0 = no effect, 1 < 20%, 2 = 20–40%, 3 = 40–60%, 4 = 60–80%, 5 = 80–90%, 6 = complete resolution).

(4) Safety: as part of our routine activity, patients and their next of kin are encouraged to take note and report at the subsequent treatment session any side effects in terms of type, duration, severity and medical intervention, if required. The severity of side effects was quantified with a score ranging from 1 to 3 (1 = mild, 2 = moderate, 3 = severe), based on the patient’s subjective experience and the medical attention required. Side effects were considered ‘severe’ when they required medical intervention (i.e., severe dysphagia, neck muscle weakness with drooping head). Side effect duration was measured using a score ranging from 1 to 3 (1 = less than one week, 2 = between one week and one month, 3 = more than one month).

Statistical analysis was performed using the Statistical Package for Social Science (SPSS), release 15.0. Continuous variables were expressed as the number of observations, mean ± standard deviation (SD), minimum and maximum value and categorical variables displayed as frequencies. We performed a one-way ANOVA analysis to compare the mean value of the efficacy parameters (i.e., latency, total duration of improvement, duration of maximum benefit, CGA) of the different types of BoNT/A. Statistical significance was considered for a p value < 0.05, and the corresponding effect size η2 was calculated.