Subjective versus objective symptom intensities ratings in cervical dystonia and hemifacial spasm across a botulinum neurotoxin cycle

Abstract Background Subjective symptom complaints often do not match the expert’s ratings in focal dystonia. Nonetheless, perceived symptom intensities drive compliance and outcome of botulinum neurotoxin (BoNT) treatment. Methods Perception of symptom development across a BoNT cycle was obtained in 21 cervical dystonia (CD) and 15 hemifacial spasm (HFS) patients at four time points during a BoNT cycle. Subjective assessments were recorded by means of a quality‐of‐life questionnaire and a patient diary containing items related to subjective severity of disease, mood, pain, social impairment, and quality of life. Medical investigation used the Tsui score and TWSTRS, and a HFS rating score, respectively. Results In both patient groups, subjective intensities were strongly associated with psychological variables. Only in CD did objective assessment moderately correlate with subjective ratings solely at the beginning and the end of the BoNT cycle. Overall, the beneficial effects of BoNT treatment were only loosely associated with subjective experiences in both groups. Conclusion The emotional situation should be assessed regularly in patients undergoing BoNT therapy.


| INTRODUC TI ON
Primary cervical dystonia (CD) is the most common form of focal dystonia. The resulting motor symptoms are the most pronounced attributes of CD and influence the degree of disability. (Ben-Shlomo et al., 2002) Due to the resulting pain and movement restrictions, CD is a highly stigmatizing and disabling condition associated with an enormous functional impairment in a patient's daily life. Hall et al., 2006;Papathansiou et al., 2001;Rinnerhaler et al., 2006;Tomic et al., 2016) CD-related bad postures, tense facial expressions, and compensatory postures lead to a socially conspicuous appearance. (Fabbrini et al., 2010;Guendel et al., 2001;Kim et al., 2016;Ozel-Kizil et al., 2008;Rinnerhaler et al., 2006) As a result, social situations are avoided frequently with harmful consequences for the satisfaction with BoNT treatment. (Leplow et al., 2017;Skogseid et al., 2007) There is rising evidence that nonmotor symptoms, such as pain, depression, or anxiety, play a pivotal role in quality of life (QoL) and treatment success. (Ben-Shlomo et al., 2002;Fabbrini et al., 2010;Jankovic & Orman, 1984;Mueller et al., 2002;Werle et al., 2014) The fact that psychiatric disorders arise before the onset of the movement disorder underlines the possibility of a shared pathophysiology. (Contarino et al., 2016;Fabbrini et al., 2010;Smit et al., 2016) The clinical picture of CD depends on not only neuropathology, but also psychological factors, which means that objective therapeutic improvements do not automatically result in corresponding improvements in a patient's QoL. (Ben-Shlomo et al., 2002;Leplow et al., 2017;Yang et al., 2017) Treatment of CD with locally injected botulinum neurotoxin (BoNT) is the first choice for the reduction in symptoms. BoNT injections are usually administered approximately at a three-month interval (BoNT cycle), with the maximum effect after four weeks and decreasing benefit within the next 6-8 weeks. The therapy effect is usually assessed by a patient's self-report and possibly using clinical scales at the end of a BoNT cycle (when BoNT is not supposed to have an effect anymore). Hence, the adaption of BoNT therapy is mainly based on patients' reports. However, the adequate assessment and comparison of subjective and objective effectiveness during a BoNT cycle are usually not feasible in clinical practice.
Thus, we compared the evolution of subjective and objective symptoms and their perception across a BoNT cycle. Thereby, we addressed the question of whether objective rating of symptom severity matches subjective experience in patients with CD compared to hemifacial spasm (HFS), a disorder of the peripheral nervous system with muscle contractions of facial muscles, which is also treated with BoNT, but characterized by a lower frequency of psychiatric comorbidities. (Fabbrini et al., 2010;Smit et al., 2016;Yang et al., 2017) 2 | PATIENTS AND ME THODS

| Study population
Twenty-one CD (19 female, age: 57.1 ± 14.9 years) and 15 HFS (13 female, 69.9 ± 13.5 years) patients were included in the study during the consultation of the outpatient clinic at the Department

| Study design
Both patient groups were examined at four time points: At t1, the BoNT injection took place (start of the BoNT cycle), t2 was assessed four weeks later (anticipated maximum effect), t3 happened two weeks before the regularly planned upcoming injection, and t4 was scheduled while BoNT reinjection was reperformed (end of the BoNT cycle). At t1, objective assessment of severity was supplemented by a semi-structured clinical interview. Furthermore, a QoL questionnaire (see below) and a self-rating scale for depression and anxiety screening (hospital anxiety and depression scale in the German version, HADS-D) were applied, respectively.

| Survey methods
Objective assessment of severity of CD was carried out using the Tsui scale and the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS). In the case of HFS, a severity score was used based on the Jankovic Rating Scale. (Jankovic & Orman, 1984) The HFS score can take values from 0 (no symptom severity) to 4 (strong symptom severity). The Craniocervical Dystonia Questionnaire (CDQ-24) was used to investigate the health-related QoL. Due to the lack of an adequate disease-specific procedure for recording the healthrelated QoL of HFS patients, the CDQ-24-adapted in formulationwas performed in this group. Higher values are associated with a poorer QoL.
Patients of both groups kept a self-developed diary for the daily recording of subjective severity of disease, mood, pain, social impairment, and restrictions in everyday activities across the entire BoNT cycle (t1-t4; Figure 1). Proven items from Leplow et al (Leplow et al., 2017) were used, which were answered on a six-level ranking scale based on the German school grade scaling system (1 = best, 6 = worst). Using this scale, perceived symptom severity was assessed by the mean score of item 3 ("How severe was your experienced disease severity today?") and 5 ("How severe did you experience involuntary movements or head deviation today?"). Pain was separately asked about by item 6 ("How strong did you perceive pain today?"). Moreover, satisfaction with BoNT therapy was investigated by item 4 ("How satisfied have you been with the effect of BoNT therapy today?").

| Statistical analysis
For statistical analysis of the two (groups)-by-four (time points of assessment) design, a sample size calculation was performed that revealed 16 patients in each group. As we wanted to detect only large effects, we chose an effect size of 0.40, a power of 0.95, and an alpha level of 0.05 (two-tailed). Scores were taken from the diary at the respective time points (t1-t4). Rank correlations (Spearman's rho) were calculated to assess the relationship between objective and subjective severity assessments. Values were indicated as mean value ± SD. Depending on the requirements for statistical procedures, either nonparametric (Mann-Whitney U tests, t tests) or parametric tests were used for group comparisons and comparisons across times of assessment (ANOVA, Friedman's tests), respectively (see results).

| RE SULTS
Clinical baseline data did not differ between groups. In CD patients, disease duration was 10.4 ± 9.2, in HFS patients 7.8 ± 4.9 years   measurements showed significant differences between most of the points of measurement, especially in CD patients (Table 1).
Only at t1 and t4 was the subjective severity assessment of CD patients moderately correlated with objective assessments (Table 2).
However, in HFS patients, there was no correlation between objective and subjective assessments at all (Table 2). By contrast, subjective intensities were strongly associated with psychosocial variables, such as mood or social impairment, for both CD and HFS patients (Table 2). All correlations showed that high subjective symptom severity was associated with reduced quality of life, enhanced experience of stigma, negative mood, reduced daily functioning, and increased social impairment, respectively. This pattern of result was seen in both patient groups.

| D ISCUSS I ON
We could show that the subjective perception of disease severity is not mainly driven by objective motor symptoms, but is primarily determined by a number of psychological variables, such as QoL in both CD and HFS (Table 2). Another core result is that the undoubt- This is what has to be expected if perception of motor function dissociates from the patient's psychological status.
Such a dissociation is highly relevant for the evaluation of therapeutic effects not only in clinical studies, but also in clinical routine, when treatment adjustments (e.g., regarding dosage or injected muscles) are made based on the patient's assessment, especially between two BoNT injections, when the patient is not routinely seen. In other words, as long as we do not combine subjective ratings and objective assessment, we do not really know whether BoNT has worked on motor symptoms during two injections.
It is well accepted that in CD (more than in HFS patients), psy- in previous studies. (Hall et al., 2006) Hence, these comorbidities have to be taken more thoroughly into consideration in the clinical management of CD and HFS patients. Objective severity of CD and HFS patients at t1 (Table 2) was in line with several other clinical studies. (Contarino et al., 2016;Kim et al., 2016) Moreover, the lack of correlation between objective symptom assessment and perceived symptom intensities perfectly fits with recent studies. (Contarino et al., 2016;Smit et al., 2016;Yang et al., 2017) All these studies showed that QoL was not primarily associated with (motor) symptom severity, but with other factors, such as pain, disability, and psychiatric burden, especially depression. (Fabbrini et al., 2010;Skogseid et al., 2007;Weiss et al., 2017)  QoL. (Junker et al., 2004;Viehmann et al., 2014) As satisfaction with BoNT therapy, compliance to medical treatment, coping behavior, and QoL depend on psychological factors more than the objective status of motor functioning, these factors need closer attention. Interestingly, this consideration seems to be valid for both CD and HFS patients. This is especially remarkable because in CD, nonmotor symptoms, especially depression, are believed to be primarily part of a shared pathophysiology. (Fabbrini et al., 2010) From BoNT studies in poststroke spasticity, also a disorder of the central nervous system, such a high impact of psychiatric comorbidities on the therapeutic effect and QoL is not known and has not been investigated systematically.
One would therefore hypothesize that subjective changes in symptom severity that are driven by mental disorders, such as depression, would correlate with objective assessment in CD rather than in HFS. However, psychiatric burden and numerous psychological variables seem to be potent general confounders or drivers of motor disease severity, which are widely independent of the specific disease. We therefore recommend that these factors be carefully considered, for example, by means of a brief structured interview addressing emotional burden, distress, impairment of quality of life, experience of stigma and its related states of symptom exacerbation, quality of life, and alterations of disease perception ( Figure 1). These questions should be combined with simple, well-anchored rating scales (see methods section) with respect to both the evaluation of clinical trials and patients' management of daily practice, at least when BoNT therapy is supposed to be adapted. Furthermore, drug treatment or psychotherapeutic interventions should be considered in patients with mental stress or illness.

CO N FLI C T O F I NTE R E S T
The authors declare no potential conflict of interests.

AUTH O R CO NTR I B UTI O N
JW acquired the data, analyzed and interpreted the data, drafted the manuscript, and approved the final version of the manuscript to be published. DW conceived and designed the manuscript, acquired the data, analyzed and interpreted the data, drafted the manuscript, critically revised the manuscript for important intellectual content, and approved the final version of the manuscript to be published. BL conceived and designed the manuscript, analyzed and interpreted the data, drafted the manuscript, critically revised the manuscript for important intellectual content, and approved the final version of the manuscript to be published.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/brb3.2023.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.