Altogether 100 consecutive ICD patients were included in this study. There were no dropouts. The patient age at ICD onset was 45.8 ± 13.5 years. The patient age at study inclusion was not used for further evaluations, as it may be confounded by the presence or absence of remissions. 66% of the patients were female, 34% were male, i.e. the female/male ratio was 2.0. The observation period was 17.5 ± 11.5 years.

After preliminary screening of the onset phase duration, patients were divided into two distinct groups as shown in Table 2. Separation was based on an onset phase of more or less than 6 months.

81% of all ICD patients belonged to this group. 67% of them were female, 33% male, i.e. their female/male ratio was 2.0. Their age at ICD onset was 46.9 ± 13.4 years. 28% had a family history of dystonia. One patient (1%) had experienced excessive psychological stress preceding ICD onset. The natural course of all ICD-type 1 patients is shown in Fig. 2. The onset phase was 3.8 ± 3.5 years (minimum 1 year, maximum 30 years). Their plateau phase was 14.9 ± 10.3 years. During this time, there was no further disease progression. In 5% of these patients, there was remission. It started 14.3 ± 8.3 years after ICD onset and had an extent of 22.5 ± 18.9% (minimum 10%, maximum 50%) of the maximal severity. All remissions (with one exception) occurred after the patient's retirement. 11% of ICD-type 1 patients had mild additional dystonia manifestations before ICD onset, including blepharospasm (6%), oromandibular dystonia (4%), writer's cramp (3%) and arm dystonia (3%). 26% of ICD-type 1 patients had mild additional dystonia manifestations after ICD onset including arm dystonia (9%), oromandibular dystonia (7%), spasmodic dysphonia (6%), blepharospasm (6%), writer's cramp (4%) and axial dystonia (3%). The observation period in ICD-type 1 was 19.1 ± 11.7 years.

The natural course of cervical dystonia-type 1 as reconstructed from data from 81 patients

19% of all ICD patients belonged to this group. Their age at ICD onset was 41.5 ± 12.7 years. 63% of ICD-type 2 patients experienced excessive psychological stress. 16% of them had a family history of dystonia, 63% of them were female, 37% male, i.e. the female/male ratio was 1.7. The natural course of each ICD-type 2 patient is shown in Fig. 3. The onset phase of ICD-type 2 patients was 8.7 ± 8.0 weeks (minimum 1 week, maximum of 24 weeks by definition). 63% of ICD-type 2 patients experienced remissions. They started 1.2 ± 0.4 years after ICD onset with an extent of 52.5 ± 34.4% (minimum 10%, maximum 100%). As in ICD-type 1 patients, the plateau phase was never progressive. 16% of ICD-type 2 patients developed mild additional dystonia manifestations after ICD onset including arm dystonia, spasmodic dysphonia and blepharospasm (no statistical analysis due to small sample sizes). One % of ICD-type 2 patients developed mild additional dystonia manifestations before ICD onset (no statistical analysis due to small sample size). None of these additional manifestations dominated the clinical picture.

The natural course of idiopathic cervical dystonia type 2 in each individual patient. A Patients with remissions. B Patients without remissions

Compared to ICD-type 1, ICD-type 2 was less frequent (19% versus 81%, binomial test (H0: p = 0.5), p < 0.001), had more often excessive psychological stress preceding ICD onset (63% versus 1%, Pearson's chi-squared test, Χ2 = 52.22; p < 0.001), more remissions (63% versus 5%, Pearson's chi-squared test, X2 = 38.81; p < 0.001) and its onset phase was shorter (8.7 ± 8.0 weeks versus 3.8 ± 3.5 years. Whether latency and extent of the remissions and prevalence of patients with additional dystonic manifestations before or after ICD onset were different in both groups, could not be decided due to limited sample size in ICD-type 1. Age at ICD onset (41.5 ± 12.7 years versus 46.9 ± 13.4 years, Student's t-test, t (98) = 1.58; p = 0.12), frequency of family history with dystonia (16% versus 28%, Pearson's chi-squared test, X2 = 1.27; p = 0.26) and female/male ratio (1.7 versus 2.0, Pearson's chi-squared test, X2 = 0.08; p = 0.77) were not significantly different.

13% of all ICD patients experienced excessive psychological stress preceding ICD onset, including partner conflicts (divorce and separation, domestic violence), special familial burdens, legal disputes and migration. In patients with excessive psychological stress, age at ICD onset was 39.0 ± 13.9 years, onset phase 0.3 ± 0.8 years, remission rate 92%, remission extent 54.5 ± 35.3%, female/male ratio 1.2.

Compared to patients without excessive psychological stress, age at onset was lower (39.0 ± 13.9 years versus 46.9 ± 13.2 years, Student's t test, t (98) = 1.99; p = 0.05), onset phase was shorter (0.3 ± 0.8 years versus 3.6 ± 3.5 years, Student's t-test, t (98) = 3.31; p = 0.001) and the remission rate was higher (92% versus 5%, Pearson's chi-squared test, Χ2 = 52.34; p < 0.001). The female/male ratio was not different (1.2 versus 2.1, Pearson's chi-squared test, Χ2 = 0.98; p = 0.32).

16% of all ICD patients experienced remissions. 75% patients with remissions had experienced excessive psychological stress, whereas only 2% of patients without remissions experienced exceptional psychological stress (Pearson's chi-squared test, Χ2 = 52.34; p < 0.001). The presence of remissions was correlated with lower patient age (55.9 ± 14.8 years vs. 65.1 ± 11.8 years, Student's t test, t (98) = 2.70; p < 0.01) and male patient sex (female/male ratio 0.8 versus 2.4, Pearson's chi-squared test, Χ2 = 4.20; p < 0.05), but not with patient age at ICD onset (versus, Student's t test, t (98) = 1.62; p = 0.11) or the presence of a family history of dystonia (23% versus 36%, Pearson's chi-squared test, Χ2 = 0.52; p = 0.47).

26% of all ICD patients had a family history of dystonia. Table 3 shows the various dystonia manifestations and their frequencies in 42 family members with history of dystonia. A family history of dystonia was not correlated with patient age at ICD onset (41.7 ± 16.8 years 47.3 11.9 years, Student's t test, t (98) = 1.96; p = 0.07), patient sex (female/male ratio 2.0 versus 1.9, Pearson's chi-squared test Χ2 = 0.01; p = 0.94), ICD-type 2 (3/26 12% versus 16/74, Pearson's chi-squared test, Χ2 = 1.27; p = 0.26) or the occurrence of remissions (13% versus 22%, Pearson's chi-squared test, Χ2 = 0.52; p = 0.47). In families with history of dystonia, an average of 1.8 ± 1.2 family members (minimum 1, maximum 5) were affected. Table 3 shows their dystonia manifestations and their relative frequencies.

97% of all patients in this study received botulinum toxin therapy. The duration of botulinum toxin therapy was 11.5 ± 9.8 years (minimum 1 year, maximum 41 years). Efficacy of botulinum toxin therapy was a subjective improvement of 73.2 ± 13.9% (minimum 30%, maximum 90%). No patient developed antibody-induced therapy failure and no patient terminated botulinum toxin therapy.