Methods

Patients

Eighty-six adult patients with PFD (>18 years) were consecutively recruited from two dystonia outpatient clinics in university hospitals in northern Germany. Refusal rate was approximately 20%. Patients with clinical evidence of secondary dystonia were excluded. The study was approved by the local ethics committees and written informed consent was obtained from all participating subjects in accordance with the Declaration of Helsinki.

Neurological examination was carried out by three experienced movement disorder specialists (JH, TB, AM). Age of onset for dystonia was determined as the time of first symptoms (eg, either dystonia or tremor) (see table 1 and supplementary material available online).

All patients tested negative for the GAG deletion in the DYT1 gene.

Psychiatric/psychological assessment

Psychiatric (axis I) and personality (axis II) disorders according to the DSM-IV criteria9 were established by two specially trained psychiatrists (HR, JS) using all available clinical data, including responses on the Structured Clinical Interview for DSM-IV (SCID).10 Diagnoses were to be agreed upon independently by two additional experienced psychiatrists (SS, RL). For diagnosis of secondary social phobia, DSM diagnostic criteria A–G had to be fulfilled, thereby clearly distinguishing social phobia from embarrassment due to dystonia symptoms.11

Whenever possible, the 5 factor inventory (NEO-FFI)12 was applied to assess the value of replicable personality traits such as neuroticism, extraversion, openness, agreeableness and conscientiousness complementary to DSM-IV personality disorders. Mean scores obtained with the NEO-FFI were compared with standardised means12 by computing effect size estimates of key measures as a descriptive approach using Cohen’s d procedure.13

Statistical analysis

Prevalence rates of DSM-IV axis I disorders in the PFD sample were compared by logistic regression analysis with those of a representative sample consisting of 3943 individuals aged 18–64 years randomly drawn from registration office files of the general population in northern Germany.14 Comparable with the SCID, the control sample was evaluated by personal interview for establishing all relevant psychiatric axis I diagnoses according to the DSM-IV (ie, the Munich Composite International Diagnostic Interview (M-CIDI)).15 Clinical reappraisal studies have shown that CIDI-SCID agreement in DSM-lV diagnoses is generally good.16 To adapt both samples, only PFD patients younger than 65 years (n = 57) were included. Sex and age were implemented as independent variables, and each of the psychiatric disorders as dependent variables to calculate odds ratios (OR) with 95% confidence intervals (CI). We used the Breslow Day Test to test for homogeneity of ORs.17

The prevalence rates for personality disorders were compared by χ² tests with those of a population based random sample based on 2053 subjects (1142 men, 911 women) from Northern Europe (18–64 years) selected for the evaluation of DSM-IV personality disorders.18

Results

Clinical characterisation

Within the group of 86 PFD patients, there was a greater proportion of women (73.3%) who were older than the men (mean 57.9 vs 50.0 years; p = 0.03). Cervical dystonia was the most frequent PFD form (81.4%). The 16 patients with blepharospasm were older (mean 64.6 vs 53.8 years; p = 0.002) and had an older age of onset (mean 54.9 vs 43.8 years; p = 0.005) than patients with cervical dystonia. Blepharospasm was more common in women (23.8%) than in men (4.3%; see table 1 in the supplementary material available online).

Psychiatric and personality disorders

Lifetime prevalence for any psychiatric or personality disorder was 70.9% (table 1). Prevalence rates of lifetime psychiatric and personality disorders did not differ between the subgroups with cervical dystonia and blepharospasm.

Compared with the population based sample, PFD patients had a 4.5-fold increased chance for any psychiatric disorder, including 25 patients with two or more psychiatric disorders. Highest chances were found for social phobia (odds ratio (OR) 21.6, 95% confidence interval (CI) 11.1 to 41.9) followed by agoraphobia (OR 16.7, 95% CI 7.4 to 37.3) and panic disorder (OR 11.5, 95% CI 4.5 to 29.2), obsessive–compulsive disorder (OR 8.4, 95% CI 1.7 to 38.9), alcohol abuse (OR 9.6, 95% CI 4.5 to 20.3) and drug dependence (OR 14.9, 95% CI 3.1 to 70.7). Chances for mood disorders were only moderately increased (OR 3.0, 95% CI 1.7 to 5.5 for major depression; see table 2 in the supplementary material available online). Most of the psychiatric disorders, except for social phobia, were present prior to the manifestation of dystonia symptoms (table 1). With respect to personality disorders, increased rates compared with the population based sample were found for cluster C disorders (32.6%), including obsessive–compulsive (22.1%) and avoidant personality disorders (16.3%; see table 3 in the supplementary material available online).

NEO-FFI personality factors

Fifty-six of the 86 PFD patients also completed the NEO-FFI (fig 1 and see table 4 in the supplementary material available online). These patients did not differ in terms of prevalence of psychiatric disorders or demographic and clinical features from the non-completers. Generally, neuroticism scores were higher in PFD patients affected with either psychiatric or personality disorders than in unaffected patients (see table 5 in the supplementary material available online). Mean openness scores in patients were decreased compared with standardised means,12 reflecting a more conventional behaviour, conservative ideas and the preference of approved settings, with subdued emotional reactions. Agreeableness scores were increased in both PFD women and men, indicating a pronounced tendency to be compassionate and cooperative. For conscientiousness, increased scores were observed without gender difference, indicating obsessive orderliness and perfectionism. Decreased neuroticism scores in PFD women indicated reduced emotional reactivity whereas PFD men scored lower on extraversion compared with the standardised group, characterising them as more introverted and autonomous.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Comparison of mean 5 factor inventory (NEO-FFI) scores in patients with primary focal dystonia (PFD), with standardised means. *Medium effect size of ⩾0.5 and <0.8 according to Cohen.13 **Large effect size of ⩾0.8 according to Cohen.13

Discussion

Distinct neuropsychiatric and personality profiles in PFD

Based on a typical PFD population,19 we confirmed previous reports on increased chances for psychiatric disorders in PFD patients, ranging from an OR of 3.0 for major depression to 21.6 for secondary social phobia. More importantly, our findings of increased rates of anxious personality disorders (cluster C) considerably expand such reports,1 2 3 4 implying that there is a continuum from distinct personality traits and certain personality disorder clusters to the manifestation of mainly anxiety disorders in PFD patients. From the personality trait perspective, PFD patients were characterised by a preference for approved settings and obsessive characteristics on the one hand and a tendency to compromise one’s interests with others but to avoid conflicts on the other. Correspondingly, more detailed analyses showed that higher neuroticism scores were found especially in those patients diagnosed with a psychiatric or personality disorder.

Although all patients were successfully treated with botulinum toxin, the chances of developing secondary social phobia were extremely high. Social phobia has been shown to be even more common in PFD than in other possibly stigmatising disorders such as alopecia areata and thus is unlikely to be a mere consequence of disfigurement.2 Development of secondary social phobia may have been facilitated by the pronounced personality traits revealed by the NEO-FFI, implying that PFD patients have difficulties in coping with dystonia symptoms and thus avoid social situations and being evaluated by others.

Increased chances for alcohol abuse and drug dependence in men may be a consequence of the fact that motor symptoms in dystonia are often relieved by alcohol, benzodiazepines or anticholinergic drugs.20 Interestingly, results from genetic studies suggest that there may be an additional neurobiological factor predisposing to substance related disorders in PFD.21 22

Neurobiological basis for motor and psychiatric symptoms in PFD

Dysfunction of the basal ganglia-thalamo-cortical circuits has been assumed to underlie both motor and psychiatric symptoms.7 The link between networks subserving mental and motor functions may be provided by direct affective input to the caudate nucleus and the thalamus originating from the amygdale and the orbitofrontal cortex.23 It is conceivable that in a disorder such as PFD, disturbed neural activity in motor loops linking the basal ganglia via the thalamus to the frontal cortex may also have an influence on the so-called limbic loops which mediate attentional, cognitive and limbic functions resulting in both altered motor and affective processing.24

Genetic factors may be another shared neurobiological basis. There is a range of inherited movement disorders where psychiatric disorders are part of the phenotypic spectrum of the genetic mutation.25 In myoclonus–dystonia, higher rates of obsessive–compulsive disorder and alcohol dependence were found in SGCE mutation carriers.21 22 This example illustrates that psychiatric disorders may be linked to the same genetic abnormality that is associated with a distinct movement disorder.

Limitations

We cannot exclude selection bias due to the fact that only PFD patients agreeing to botulinum toxin treatment were recruited and that they may have scored higher on agreeableness than those who refused, which is a general limitation of clinical studies. Based on the assumption that different dystonia subtypes share common aetiological factors,26 we did not intend to make any statement with respect to specific neuropsychiatric features of specific PFD subgroups or differences between PFD subgroups. In fact, a high phenotypic variability has been observed even within families with multiple cases of dystonia. Further studies with larger sample sizes of subgroups may be able to answer this question.

In conclusion, the results of the present study expand earlier findings showing more specifically that anxiety spectrum disorders represent part of the phenotypic PFD spectrum. Interdisciplinary research effort is needed to unravel the shared disease mechanisms of both motor and neuropsychiatric symptoms in PFD with high resolution imaging and genetic approaches being the most promising. Distinct symptoms may be associated with specific alterations of limbic and frontal circuitry.27