GRAPHICAL ABSTRACT

INTRODUCTION

Currently, botulinum toxin type A (BTX-A) injection is the primary treatment for cervical dystonia (CD) [1]. However, individual responses to BTX-A vary significantly [2], making optimal results difficult to achieve consistently. The variability in response is largely influenced by selecting the appropriate dystonic muscles, which is essential for maximizing the effectiveness of BTX-A treatment while minimizing adverse effects [3].

To address the complexity of CD, Reichel [4] introduced the Col-Cap classification, which is based on head and/or neck movements in the coronal, sagittal, or transverse plane. This classification categorizes CD postures into torti-, latero-, ante-, and retro-collis/caput patterns, as well as shift forms. Although this concept is intriguing, the most commonly injected muscles appear to have remained constant over the past decades [5], and evidence of improved outcomes following its application in clinical practice is limited [6-9]. Furthermore, the efficacy of BTX-A in these various CD patterns remains unclear.

Recently, an innovative approach that uses ^99mTc-sestamibi single-photon emission computed tomography (^99mTc-MIBI SPECT) was developed to better identify dystonic muscles in CD patients. A randomized clinical study revealed that the combination of SPECT with electromyography (EMG) for identifying dystonic muscles can increase the clinical efficacy of BTX-A [10]. However, this previous study focused on the overall efficacy of BTX-A across various types of CD, and whether clinical improvement rates differ among different CD types when employing SPECT combined with EMG is unclear.

Thus, the aim of this study was to evaluate and compare the clinical improvement rates among patients with different CD types after BTX-A injection using SPECT combined with EMG. Furthermore, we sought to systematically identify dystonic target muscles in diverse CD types on the basis of SPECT and EMG findings to inform and guide clinical treatment.

MATERIALS AND METHODS

Study subjects

This retrospective study included patients with CD who visited the neurology outpatient department at Shanghai Tongji Hospital from September 2017 to December 2022. The study was approved by the Institutional Review Board of Tongji Hospital, Shanghai (Approval No. 2021-013). Written informed consent was obtained from the parents or legal guardians of each participant.

The inclusion criteria were as follows: 1) age between 18 and 75 years, irrespective of sex; 2) diagnosis of idiopathic CD; 3) duration of CD symptoms of at least one month; 4) no history of surgical treatment for CD; and 5) completion of ^99mTc-MIBI SPECT before BTX-A injection. The exclusion criteria were as follows: 1) congenital torticollis or abnormal head–neck posture caused by tumors, trauma, medication, or other factors and 2) idiopathic CD with an isolated head tremor.

CD pattern evaluation

At our center, standardized video recordings were obtained for each CD patient from both frontal and lateral perspectives, capturing seated, standing, and stepping positions. Two experienced neurologists independently reviewed these recordings and classified dystonic patterns according to the concept of Col-Cap prior to BTX-A administration. In instances of interrater disagreement, a third senior neurologist was consulted to adjudicate the classification through re-evaluation, and the final pattern was determined by majority consensus. According to the Col-Cap classification model, the C2 plane serves as a demarcation line. Patterns above the C2 plane are categorized as abnormalities of the head, such as torticaput, laterocaput, anterocaput, and retrocaput, whereas patterns below this plane are classified as neck abnormalities, including torticollis, laterocollis, anterocollis, and retrocollis, along with categorization into sagittal shift and lateral shift on the basis of the relative displacement between the central axis of the head and trunk. The presence of tremors in CD patients was assessed, and tremor patterns were documented as “no-no,” “yes-yes,” or “multidirectional.” A single CD pattern was defined as the simple type, whereas a combination of two or more overlapping CD patterns was classified as the complex type.

^99mTc-MIBI SPECT examination

^99mTc-MIBI (Shanghai Atomic Kexing Pharmaceutical Company Limited) was used as the developing agent. SPECT scans were performed with Precedence SPECT (Philips) one hour after the intravenous injection of 740 MBq of ^99mTc-MIBI. During this one-hour interval, patients were instructed to minimize voluntary muscle contractions and avoid attempting to correct abnormal postures to reduce compensatory muscle contractions. Computed tomography (CT) was performed with a slice thickness of 5 mm, a pitch of 1, and a 512×512 matrix. SPECT was subsequently performed, with a peak acquisition energy of 140 keV, window width of 20%, matrix size of 64×64, magnification of 1, probe rotations of 360° at 6° per frame, with a frame duration of 15 s and a total of 32 continuous frames. After acquisition, the SPECT and CT images were fused [11,12].

The images were assessed by a skilled nuclear medicine physician. Muscles exhibiting abnormal uptake of ^99mTc-MIBI were identified as dystonic muscles. The muscle/background ratio (MBR) for each dystonic muscle was subsequently calculated as described previously [12].

Dystonic muscle selection and BTX-A injection

Prior to the BTX-A injection, an experienced neurologist reassessed and confirmed the CD pattern, ensuring consistency with prior evaluations. Subsequently, the treating physician identified candidate muscles after clinical evaluation and viewing the SPECT/CT images. During the injection procedure, patients were assessed in both the resting and functional positions using EMG for each candidate muscle. Injections were performed if the EMG results were positive. If a patient had negative EMG results across all positions but a positive SPECT result, we prioritized the EMG findings and did not proceed with the injection.

A 0.5×60 mm needle (Suzhou Haishen Co., Ltd.) was employed for the injections, guided by EMG. Every CD patient was given the option to choose the type of botulinum toxin according to their preference, between Hengli (Lanzhou Biologicals Co., Ltd.) and Botox (Allergan), which had been proven to have a dose ratio of 1:1 and the same therapeutic effects [13-15]. The dose for each muscle was mainly based on clinical evaluation, clinical experience, and muscle volume. EMG amplitude and MBR on SPECT were also considered references for dose selection, especially for deep muscles.

Data collection

The baseline characteristics of the participants were retrieved from their medical records. The frequency of injection (%) for each cervical muscle was recorded. This frequency was determined as the ratio of the number of CD patients injected in a specific muscle to the total number of CD patients with the same dystonic pattern. Concurrently, the dosage for each injected muscle was also recorded. The efficacy of BTX-A treatment was assessed based on the patient’s subjective improvement from baseline (prior to treatment) at four weeks post-injection, expressed as a percentage. This improvement was scored on a visual analog scale ranging from 0 to 100. The Tsui score at four weeks post-injection was evaluated by an experienced neurologist, and the reduction in the Tsui score was also calculated.

Statistical analysis

Quantitative variables are expressed as the mean±standard deviation. Categorical variables are presented as counts (percentages) and were analyzed using chi-square tests or Fisher’s exact tests, as appropriate. The Kolmogorov–Smirnov test was employed to assess the normality of continuous data. Independent sample t-tests or Mann–Whitney U tests were used for normally and nonnormally distributed data, respectively. The Kruskal–Wallis H test was applied for comparisons across multiple groups. A p-value of less than 0.05 was considered to indicate statistical significance. The IBM SPSS 22.0 software package was used for statistical analysis, and GraphPad Prism Software version 8.0.1 was used for statistical graph creation.

RESULTS

Incidence of different single subtypes and different combinations of CD

A total of 252 patients with CD were enrolled, and the baseline characteristics of the study participants are shown in Table 1. Patients with only one abnormal movement pattern were classified as having simple CD, whereas patients with two or more abnormal movement patterns were considered to have complex CD. We noticed that patients with complex type (69.0%) outweighed patients with the simple type (31.0%) in the cohort.

Among the patients with simple CD, torticaput (34.6%) was the most common pattern, followed by laterocaput (33.3%) and retrocaput (14.1%), whereas anterocaput (5.1%) and anterocollis (3.8%) were less common (Table 2). For patients with complex CD, combinations of two forms were most common (74.1%). The most common combination was either torticaput combined with another pattern or laterocaput combined with another pattern, with torticaput + torticollis being the most common pattern (58.1%). Patients with combinations of three simple forms were relatively rare (20.1%), with torticaput + torticollis + laterocaput (31.4%) being the most common pattern, followed by torticaput + torticollis + retrocaput (28.6%). Patients with four patterns were very rare (5.7%), with most patients combining a fourth type with the torticaput + torticollis + laterocaput types (Table 2).

Among the 252 CD patients, 73 (29.0%) exhibited tremors, with all three types of tremors observed. Most patients (67.1%) had “no-no” tremors, with 18 patients (24.7%) having “yes-yes” tremors and six patients (8.2%) having “multidirectional” tremors. Among all patients with tremors, torticaput (68.5%) was the most common dystonic subtype associated with tremors. Patients with “no-no” tremors often had complex cervical movement patterns (65.3%), with torticaput + torticollis (32.7%) being the most common. In patients with “yes-yes” tremors and “multidirectional” tremors, both the simple type and the complex type were equally common, without any particularly prominent pattern.

Clinical improvement rates after BTX-A injection in different CD subtypes

In our study, 72 CD patients (28.6%) received Botox (Allergan), whereas the remaining patients (71.4%) chose Hengli (Lanzhou Biologicals Co., Ltd.). The subjective improvement rates at four weeks after injection for all 252 patients were recorded. However, only 209 patients had Tsui scores assessed at four weeks post-injection. Owing to missing follow-up data of less than 20%, the reduction in the Tsui score was also used to evaluate the efficacy of BTX-A.

We first compared patients with simple and complex CD. Neither sex nor age significantly differed between the two groups. Clinical severity, as measured by the Tsui score, was significantly greater in the complex-type group (10.6±3.9) than in the simple-type group (8.4±3.0, p<0.001) (Figure 1A). Under the guidance of SPECT and EMG, the BTX-A dose used to treat patients with complex CD (264.9±52.0 units) was compared with that used to treat patients with simple CD (254.8±59.4 units; p=0.228) (Figure 1B). However, the subjective improvement rates after BTX-A injection were not significantly different between the simple-type group (63.5%±20.6%) and the complex-type group (59.9%±22.2%, p=0.247) (Figure 1C). Similarly, the reduction in the Tsui score did not significantly differ between the simple-type group (52.4%±24.2%) and the complex-type group (49.6%±23.6%, p=0.556) (Figure 1D).

We subsequently compared patients with different simple subtypes. Because some subtypes were underrepresented in our cohort, we included only subtypes with at least five cases in the analysis. A total of four simple subtypes were analyzed, as depicted in Figure 1. The clinical severity as indicated by the Tsui score (p=0.671) (Figure 1E), BTX-A dose (p=0.121) (Figure 1F), subjective improvement rate (p=0.650) (Figure 1G), or reduction rate of the Tsui score (p=0.374) (Figure 1H) did not significantly differ among these subtypes.

We further compared complex-type patients with different combinations of subtypes. Similarly, we included only combinations with at least five cases in the analysis. Seven different combinations were analyzed, as illustrated in Figure 1. Owing to the small number of patients in some groups, the groups were not directly compared, but we still observed that the Tsui scores seemed to be higher in patients with three forms of CD than in those with two forms of CD (Figure 1I). However, the BTX-A dose did not increase with increasing disease severity (p=0.311) (Figure 1J). Additionally, no significant difference was observed in the subjective improvement rates (p=0.620) (Figure 1K) or the reduction rates of the Tsui score (p=0.357) (Figure 1L) among the various complex patterns.

Clinical improvement rates after BTX-A injection in patients with tremors

We first compared CD patients with and without tremors. Sex, age, or duration of symptoms did not significantly differ between the two groups. Although the clinical severity in the tremor group (11.6±3.6) was greater than that in the group without tremors (9.3±3.6; p<0.001) (Figure 2A), the BTX-A dose (p=0.332) (Figure 2B), subjective improvement rate (p=0.282) (Figure 2C), and reduction rate of the Tsui score (p=0.227) (Figure 2D) after BTX treatment did not significantly differ between the groups.

Comparisons among CD patients with “no-no” tremors, “yes-yes” tremors, and “multidirectional” tremors were subsequently performed. Clinical severity (Tsui score) did not significantly differ among the three groups (p=0.364) (Figure 2E). Moreover, the BTX-A dose (p=0.253) (Figure 2F), subjective improvement rate (p=0.083) (Figure 2G), or reduction rate in the Tsui score (p=0.470) (Figure 2H) did not significantly differ.

Dystonic muscles in different CD subtypes

To better guide future clinical practice, we analyzed the dystonic muscles of different subtypes on the basis of the SPECT and EMG results in detail. Because complex subtypes are characterized by an overlay of various simple types, analyzing their muscle excitation characteristics has limited value for clinical practice. Thus, we focused primarily on the dystonic target muscles in simple types (only patterns with at least 10 cases were analyzed) (Table 3). However, the torticaput + torticollis pattern was also included in the analysis because of its high prevalence and relatively consistent muscle combination pattern.

A total of five major CD subtypes were analyzed in this study. The dystonic muscles targeted for each pattern, along with their respective injection frequencies, are presented in Table 3. Additionally, the corresponding BTX-A injection doses for different dystonic muscles are also provided.

Adverse events

All adverse events were related to BTX-A injection (Table 4). Local pain at the injection site, cervical weakness, and dry mouth were commonly observed, whereas dizziness, dysphagia, and influenza-like syndrome were rare. All adverse events were very mild, with no special treatment needed.

DISCUSSION

The present study represents the first evaluation of clinical improvement rates among different CD types evaluated using the Col-Cap classification following BTX-A injection. By employing multifaceted approaches to identify the dystonic muscles, we discovered that the efficacy of BTX-A treatment was equivalent in patients with different types of CD, and we further elucidated combinations of dystonic muscles in five CD types for clinical BTX-A treatment.

In terms of the Col-Cap classification, torticaput (49.0%) was reported as the most common subtype among all CD patients [4,16]. However, torticaput accounted for 34.6% of subtypes among our patients with simple patterns and only 10.7% of subtypes among all CD patients. This discrepancy may be explained by two factors. First, most CD patients who underwent SPECT at our center had previously responded suboptimally to BTX-A or had complex types of CD, and patients with a pure tortiform pattern had a relatively fixed combination of dystonic muscles and achieved better treatment outcomes with no need for additional SPECT examination; thus, this subset of patients was not included in the study. Second, we did not observe patients with pure torticollis pattern and found that all patients with a torticollis pattern also had a torticaput pattern. Therefore, the complex group included a significant portion of patients with a torticaput pattern. Our findings align with those of previous studies in which a torticollis pattern was less common, and most patients had a combination of torticaput and torticollis patterns [6].

In line with the Col-Cap classification, laterocaput (16.7%) was the second most common subtype, and 9.8% of CD patients presented with the laterocollis subtype [16]. However, the prevalence of laterocaput (10.3%) and laterocollis (2.8%) was lower than that reported by Jost [16], which might be attributed to the selection basis of more complex patients in the present study, as mentioned above. In Jost’s study, anterocaput, anterocollis, retrocaput, and retrocollis were each identified in less than 5% of the study population. The rarity of anterocaput (1.6%), anterocollis (1.2%), and retrocaput (4.4%) in our study was comparable to the previous study. The retrocollis subtype may be less prominent in clinical settings, as patients often adapt to the abnormal head posture by utilizing compensatory head flexion to alleviate discomfort or improve visual alignment. This adaptive mechanism likely accounts for the lower frequency of the retrocollis subtype observed in our study.

Reichel’s Col-Cap classification [4] identifies lateral shift as a combination of laterocollis and contralateral laterocaput patterns, and the prevalence of lateral shift (7.9%) in our study was comparable to that (9.2%) reported in a previous study [16]. In our patients with complex forms of CD, in addition to the torticaput + torticollis combination, patients who presented rotation with lateral flexion or combined the former patterns with retro/ante-flexion were the most frequently observed. Reichel reported that the majority of CD patients exhibited a combination of two or more forms, with lateral flexion coupled with rotation emerging as the most common combination [4,17], which was consistent with our study.

Furthermore, our study identified tremors in 29.0% of CD patients, with “no-no” tremors (67.1%) being the most common. In the literature, the incidence of tremors in patients with CD ranges from 18.3% to 63.0% [18-20], with “no-no” tremors also being the most commonly observed type [18]. Notably, our study revealed that patients with complex-type CD were more prone to tremors, which is consistent with the findings of previous studies in which tremors were the common clinical presentations of CD involving two or more types [21,22]. Additionally, Pandey et al. [7] reported that torticaput (57.3%) was the most common dystonic subtype associated with tremors, which was comparable to our findings (68.5%).

In this study, the mean BTX-A dose in the complex CD group (264.9±52.0 units) was comparable to that in the simple CD group (254.8±59.4 units), and no significant difference was noted among the various subtypes. The mean BTX-A dose in our study was higher than that in some previous studies (187.0±76.5 units – 189.8±87.1 units) [23,24] but comparable to the mean dose of 241.2 units (range, 95.0–360.0 units) reported in another clinical study [25]. This discrepancy might be attributable to our selection of dystonic muscles by SPECT, leading to a more precise and rational administration of BTX-A into dystonic muscles. Most of the increased doses were injected into the deep cervical muscles, yet these doses did not lead to any additional adverse events.

In the following section, the clinical improvement of different CD patients after BTX-A treatment is discussed. According to systematic reviews in the Cochrane Database, the mean subjective improvement rate among patients with various types of CD following BTX-A treatment was reported to be 55.3% [26]. The subjective improvement rate was 64.1%±20.8% for patients with simple-type CD and 58.8%±22.2% for those with complex-type CD in our study. These findings indicate that the efficacy of BTX-A was not solely determined by the complexity of CD patterns, but the accurate selection of dystonic muscles could promote CD patients to achieve satisfactory efficacy even in complex subtypes.

In our research, the subjective improvement rate for patients with the torticaput subtype was 65.6%±23.7%, which was slightly lower than that in a previous study [27]. Notably, the CD patients with tortiform subtypes included in our study were predominantly those with suboptimal clinical efficacy in previous treatments. The subjective improvement rates for patients with the laterocaput and laterocollis subtypes were 61.0%±17.6% and 57.1%±20.4%, respectively, which are comparable to those in earlier research.

Interestingly, in contrast to previous conclusions [27] suggesting greater benefits from treatment in patients with neck subtypes (torticollis or laterocollis subtypes), our study revealed smaller mean improvements in these neck subtypes than in head subtypes (torticaput or laterocaput subtypes). However, the evaluation in the study by Jost et al. [27] was conducted at five months post-treatment, whereas our evaluation occurred at four weeks, making direct comparisons between the two studies inappropriate. Nonetheless, we support the opinion that the treatment of caput subtypes tends to yield better results than that of collis subtypes.

Our study revealed a subjective improvement rate of 61.4%±22.5% for patients with retrocaput patterns, whereas the rates ranged from 52.1%±28.3% to 56.0%±20.0% when retrocaput was combined with other subtypes. A previous study [28] reported excellent relief in 24.5%, moderate relief in 32.1%, mild relief in 16.9%, and no response in 24.5% of patients with retroform CD. However, our results were slightly inferior to those of another retrospective study we conducted, with an average symptom relief rate of 69.0%±16.7% [29].

Previously, lateral shift was conventionally perceived as more resistant to BTX-A treatment [5,30]. Surprisingly, in our study, the subjective improvement rate for patients with a lateral shift was 63.5%±23.5%, and their Tsui score reduction rate was 43.2%±27.6%, comparable to or even superior to those observed in patients with the laterocollis or laterocaput subtype.

The subjective improvement rate and reduction rate in the Tsui score for patients with the forward sagittal shift pattern were 71.3%±16.0% and 62.9%±15.4%, respectively. Similar findings were echoed in another clinical investigation [31], where seven CD patients with a forward sagittal shift responded well to BTX-A injections, with a mean subjective improvement rate of 73%, owing to the injection of the suboccipital muscles and longus collis under the guidance of EMG.

Furthermore, this study compared CD patients with and without tremors. Because the Tsui scale inherently incorporates tremor severity, we observed that the Tsui score in the tremor group was greater than that in the group without tremor. However, no significant difference in the BTX-A dose or clinical improvement rate was found between the two groups. Previous studies have suggested that the dystonic muscles involved in “yes-yes” or “multidirectional” tremors are more complex than those involved in “no-no” tremors, potentially leading to worse clinical improvement rates [32]. However, in our study, CD patients with “yes-yes” or “multidirectional” tremors did not exhibit inferior improvement rates compared with those with “no-no” tremors. This finding might be attributed to the efficacy of SPECT combined with EMG in precisely exploring dystonic muscles. On the other hand, the fewer instances of CD patients exhibiting “yes-yes” or “multidirectional” tremors might have introduced a degree of bias. Therefore, further studies are needed to verify these aspects.

Our study analyzed the injection frequency and BTX-A dose of dystonic muscles in five major CD patterns, which were targeted via clinical pattern assessment, SPECT imaging, and EMG. Although we previously stated that target muscles for injection were primarily selected on the basis of concordant positive findings from both EMG and SPECT, in cases of discrepancy, greater weight was given to the EMG results. However, our previous research demonstrated a high level of concordance between muscles identified as positive by SPECT and those confirmed by EMG, supporting the reliability of using SPECT as a complementary tool in muscle selection. Previous studies [17,18,33] have suggested that the principal dystonic muscles involved in the torticaput subtype include the contralateral sternocleidomastoid (SCM), descending part of the trapezius (TPZ), ipsilateral splenius capitis (SPCa), obliquus capitis inferior (OCI), and longissimus capitis (LGCa). Compared with previous studies, our study revealed that the OCI was also the most frequently affected muscle in patients with a torticaput pattern, which is consistent with the results of our previous study [34]. Additionally, our study revealed that the ipsilateral levator scapulae (LS), scalenus (Sc), and semispinalis capitis (SSCa) also play an important role in the torticaput subtype.

The fact that LS and Sc were also considered target muscles for the torticaput subtype may be surprising because head rotation occurs between C1 and C2 in these patients according to the Col-Cap concept; thus, only the muscles acting on the skull should be affected, whereas LS and Sc act at the level below C2. However, 92.6% of our patients with the torticaput subtype had LS excitation, and 74.1% had Sc excitation, which was confirmed by both SPECT and EMG. We observed Sc excitation in a considerable proportion of patients with head rotation, and these patients reported notable improvement following injection for Sc. Although this observation is challenging to explain purely from an anatomical perspective, it highlights the importance of multidimensional assessment in identifying and targeting dystonic muscles. We acknowledge that these results require further validation and encourage future studies to explore and verify these findings.

With respect to the torticaput + torticollis combination, our study revealed the involvement of the splenius cervicis (SPCe) and that the Sc was more often injected in these patients than in patients with the torticaput subtype. The SPCe muscle functions in laterally flexing and rotating the head and neck to the ipsilateral side, and the Sc muscle plays an active role in neck movements. Therefore, when treating patients with the torticaput + torticollis pattern, physicians might also pay attention to the SPCe and Sc, which were not previously addressed in patients who only show a torticollis pattern [17].

For the laterocaput subtype, the primary dystonic muscles were relatively clear and include the ipsilateral SCM, TPZ (descending part), SPCa, SSCa, LS, and LGCa [17,33]. Additionally, our study involved injections into the ipsilateral Sc for patients with laterocaput CD. The middle and posterior portions of the Sc anatomically function in neck flexion and lateral bending of both the neck and head to the same side, suggesting their potential role in tilt forms of CD. Although some have proposed that Sc involvement is limited to neck tilt rather than head tilt, we believe that in certain cases, subclinical neck tilt may also contribute to head tilt. Additionally, we also revealed that the clavicular head of SCM might play an important role in laterocaput CD.

For retrocaput CD, previous research suggested that the dystonic muscles included the bilateral OCI, SSCa, TPZ, and SPCa [18,33]. Compared with previous studies, our team additionally injected the LS, rectus capitis posterior major (RCPM), and SCM, whereas the OCI was less commonly injected. Bilateral contraction of the LS contributes to retraction of the head and neck, whereas bilateral contraction of the SCM can lead to both head flexion and extension. However, the effect of bilateral SCM contraction depends on the initial head position: in a neutral position, it induces an anterocaput pattern, whereas in a flexed position, it may exacerbate an existing retrocaput pattern. A recent study highlighted the primary function of the RCPM in stabilizing the occipito-atlantal joint [35]. Despite the OCI being proposed as a primary muscle for retrocaput CD in some studies [18,33], we only injected this muscle in 18.2% of patients. This finding could be attributed to SPECT revealing the activation of the RCPM in retrocaput CD, which prompted additional EMG and injection in the RCPM, not the OCI.

In the context of lateral shift, the implicated muscles included the contralateral SCM, TPZ, SPCa, SSCa, and LS and the ipsilateral LS, semispinalis cervicis (SSCe), and Sc [33]. A retrospective study [30] reported BTX-A injection frequencies for selected muscles in 11 patients with lateral shifts as follows: SCM (90.9%), SPCa (63.6%), LS (36.4%), and TPZ (36.4%). Our study revealed additional injections in the ipsilateral SCM, SPCa, Sc, TPZ, and contralateral Sc, with greater injection frequencies for the SPCa (90.0%), LS (85.0%), and TPZ (50.0%) than those documented in previous studies. Lateral shift is defined as laterocollis CD associated with contralateral laterocaput CD. However, in clinical practice, distinguishing whether an ipsilateral laterocollis pattern is accompanied by a slight ipsilateral laterocaput or if the contralateral laterocaput pattern is more prominent, rendering the ipsilateral laterocaput less apparent, is difficult. Our SPECT and EMG results supported this possibility, as both techniques revealed the involvement of muscles typically associated with ipsilateral laterocaput CD, such as the ipsilateral SCM and SPCa. With respect to the contralateral Sc, we have previously explained its potential role in contralateral laterocaput CD. However, further clinical studies are needed to validate these findings and provide a more definitive understanding.

Dressler et al. [36] provided a comprehensive summary of BTX-A injection frequencies pertaining to the dystonic muscles of all CD types, including the SCM (52.0%), TPZ (78.0%), SPCa (75.0%), LS (31.0%), Sc (29.0%), and deep muscles (3.0%). Our study demonstrated higher BTX-A injection rates for both superficial and deep cervical muscles than previous findings did, underscoring the increased efficacy of the BTX-A injection approach, which utilized clinical pattern assessment, SPECT imaging analysis, and needle EMG for accurate screening of dystonic muscles.

This study has several limitations. First, our study had selection bias, as it included only those CD patients who underwent SPECT. Consequently, individuals with favorable treatment outcomes or simpler CD types, for whom SPECT was deemed unnecessary, were not included in our analysis. Nevertheless, CD patients with complex CD patterns or poor prior treatment efficacy still achieved favorable outcomes, which further supported our conclusions. Second, because this study was retrospective, most CD patients did not undergo long-term follow-up; therefore, long-term outcomes may not have been fully captured. Third, although low-dose ^99mTc-MIBI is very safe for both CD patients and the public [37,38], it is still associated with radiation exposure. Thus, we believe that, rather than for routine clinical practice, this technique is more suitable for patients with complex CD subtypes, nonresponders, and patients who develop a change in pattern. Fourth, the Tsui scale, while widely used, may not fully capture the complexity of CD, particularly when the advanced Col-Cap classification is considered. As our study was retrospective, only a subset of patients had complete Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) data, limiting our ability to perform a comprehensive analysis. We recommend the use of more comprehensive scales, such as the TWSTRS, and the incorporation of multiple outcome measures in future studies. Fifth, our previous study revealed that anatomical localization combined with EMG guidance was sufficiently precise for muscle targeting. Thus, BTX-A injections in our study were not performed under ultrasound (US) guidance. However, we still recommend that future studies incorporate US-guided injections to enhance clinical applicability and injection accuracy, particularly for beginners. Finally, another limitation of this study lies in the uneven sample size distribution across different CD subtypes. CD exhibits a wide range of clinical patterns, and many patients present with various forms involving overlapping components. As a result, the number of patients within certain specific subgroups was inevitably small, which limited the statistical power for quantitative comparisons. Therefore, subgroup analyses should be interpreted as exploratory rather than confirmatory. Future studies with larger, prospectively balanced samples are needed to further validate these findings.

In conclusion, the efficacy of BTX-A was comparable across CD patterns because of the appropriate selection of dystonic muscles aided by SPECT and EMG, indicating that employing multifaceted approaches to identify dystonic muscles can yield favorable treatment outcomes, even in complex cases.

Notes

Conflicts of Interest

The authors have no financial conflicts of interest.

Funding Statement

This work was supported by the following grants: National Key R&D Program of China (2023YFC3604500/2023Y FC3604505); National Natural Science Foundation of China (Project 32030047); Clinical Technology Innovation Project of Shanghai Shenkang Hospital Development Center (SHDC12020119); Science and Technology Innovation Program of Shanghai Municipal Science and Technology (Nos. 22Y31900200 and 22Y31900203); National Clinical Key Specialty Construction Project of China (Z155080000004); Shanghai Rehabilitation Medical Research Center (Top Priority Research Center of Shanghai) (2023ZZ02027); Shanghai Clinical Research Ward (SHDC2023CRW018B); and Shanghai Hospital Development Center Foundation— Shanghai Municipal Hospital Rehabilitation Medicine Specialty Alliance (SHDC22023304).

Acknowledgments

None

Author Contributions

Conceptualization: Yougui Pan, Hongkai Gu, Fei Teng. Formal analysis: Hongkai Gu, Fei Teng. Investigation: Hongkai Gu, Yougui Pan. Methodology: Hongkai Gu, Yougui Pan. Project administration: Lingjing Jin. Resources: Xiaolong Zhang, Lizhen Pan, Ronghua Hong, Zhuang Wu, Yougui Pan. Software: Shuzhen Chen, Chenghong Wang. Supervision: Lingjing Jin. Visualization: Fei Teng. Writing—original draft: Hongkai Gu. Writing—review & editing: Fei Teng.

Figure 1.

Comparison of baseline Tusi score, BTX-A dose, and clinical improvement rates among different CD subtypes. A: Comparison of baseline Tsui score between simple type and complex type. B: Comparison of BTX-A doses between simple type and complex type. C: Comparison of subjective improvement rates between simple type and complex type. D: comparison of reduction rates of Tsui score between simple type and complex type. E: Comparison of baseline Tsui score among four different simple patterns. F: Comparison of BTX-A doses among four different simple patterns. G: Comparison of subjective improvement rates among four different simple patterns. H: Comparison of reduction rates of Tsui score among four different simple patterns. I: Comparison of baseline Tsui score among seven different complex patterns. J: Comparison of BTX-A doses among seven different complex patterns. K: Comparison of subjective improvement rates among seven different complex patterns. L: Comparison of reduction rates of Tsui score among seven different complex patterns. The numbers in the bar chart represent the number of patients in each group. BTX-A, botulinum toxin type A; CD, cervical dystonia; ns, not significant. *p<0.05; ***p<0.001.

Figure 2.

Comparison among CD patients with different tremor types. A: Comparison of baseline Tsui score between CD with tremor and without tremor. B: Comparison of BTX-A doses between CD with tremor and without tremor. C: Comparison of subjective improvement rates between CD with tremor and without tremor. D: Comparison of reduction rates of Tsui score between CD with tremor and without tremor. E: Comparison of baseline Tsui score among CD with three different tremor types. F: Comparison of BTX-A doses among CD with three different tremor types. G: Comparison of subjective improvement rates among CD with three different tremor types. H: Comparison of reductions rates of Tsui score among CD with three different tremor types. BTX-A, botulinum toxin type A; CD, cervical dystonia; ns, not significant. ***p<0.001.

REFERENCES

• 1. Jinnah HA. Medical and surgical treatments for dystonia. Neurol Clin 2020;38:325–348.ArticlePubMedPMC
• 2. Bledsoe IO, Comella CL. Botulinum toxin treatment of cervical dystonia. Semin Neurol 2016;36:47–53.ArticlePubMed
• 3. Camargo CH, Cattai L, Teive HA. Pain relief in cervical dystonia with botulinum toxin treatment. Toxins (Basel) 2015;7:2321–2335.ArticlePubMedPMC
• 4. Reichel G. Cervical dystonia: a new phenomenological classification for botulinum toxin therapy. Basal Ganglia 2011;1:5–12.Article
• 5. Erro R, Picillo M, Pellecchia MT, Barone P. Improving the efficacy of botulinum toxin for cervical dystonia: a scoping review. Toxins (Basel) 2023;15:391.ArticlePubMedPMC
• 6. Jost WH. Torticaput versus torticollis: clinical effects with modified classification and muscle selection. Tremor Other Hyperkinet Mov (N Y) 2019 Jul 12:9 [Epub]. https://doi.org/10.7916/tohm.v0.647. Article
• 7. Pandey S, Kreisler A, Drużdż A, Biering-Sørensen B, Sławek J, Tatu L, et al. Tremor in idiopathic cervical dystonia - possible implications for botulinum toxin treatment considering the Col-Cap classification. Tremor Other Hyperkinet Mov (N Y) 2020;10:13.ArticlePubMedPMC
• 8. Kreisler A, Gerrebout C, Defebvre L, Demondion X. Accuracy of non-guided versus ultrasound-guided injections in cervical muscles: a cadaver study. J Neurol 2021;268:1894–1902.ArticlePubMedPDF
• 9. Szabó M, DO Kiem D, Gárdián G, Szpisjak L, Salamon A, Klivényi P, et al. Clinical features of cervical dystonia patients classified by the COLCAP concept and treated with ultrasound-guided botulinum neurotoxin. Ideggyogy Sz 2023;76:37–45.ArticlePubMed
• 10. Teng F, Malam Djibo I, Chen S, Su J, Pan Y, Zhang X, et al. Efficacy of single-photon emission computed tomography aided botulinum toxin injection in cervical dystonia: a double-blind, randomized study. Parkinsonism Relat Disord 2021;91:77–82.ArticlePubMed
• 11. Feng L, Zhang Z, Malam Djibo I, Chen S, Li B, Pan Y, et al. The efficacy of single-photon emission computed tomography in identifying dystonic muscles in cervical dystonia. Nucl Med Commun 2020;41:651–658.ArticlePubMed
• 12. Chen S, Issa MD, Wang C, Feng L, Teng F, Li B, et al. [99mTc]MIBI SPECT/CT for identifying dystonic muscles in patients with primary cervical dystonia. Mol Imaging Biol 2020;22:1054–1061.ArticlePubMedPDF
• 13. Barbosa PM, Rodrigues GR, de Oliveira DS, de Souza CP, Tumas V. Comparison between dysport and prosigne in the treatment of cervical dystonia. Clin Neuropharmacol 2015;38:221–226.ArticlePubMed
• 14. Feng Y, Liu W, Pan L, Jiang C, Zhang C, Lu Y, et al. Comparison of neurotoxic potency between a novel chinbotulinumtoxinA with onabotulinumtoxinA, incobotulinumtoxinA and lanbotulinumtoxinA in rats. Drug Des Devel Ther 2017;11:1927–1939.ArticlePubMedPMCPDF
• 15. Dressler D, Pan L, Su J, Teng F, Jin L. Lantox-the Chinese botulinum toxin drug-complete English bibliography and comprehensive formalised literature review. Toxins (Basel) 2021;13:370.ArticlePubMedPMC
• 16. Jost WH, Tatu L, Pandey S, Sławek J, Drużdż A, Biering-Sørensen B, et al. Frequency of different subtypes of cervical dystonia: a prospective multicenter study according to Col-Cap concept. J Neural Transm (Vienna) 2020;127:45–50.ArticlePubMedPDF
• 17. Reichel G. Dystonias of the neck: clinico-radiologic correlations. In: Rosales RL. Dystonia - The Many Facets. Online: InTech. 2012;17-32. Article
• 18. Camargo CH, Teive HA, Becker N, Baran MH, Scola RH, Werneck LC. Cervical dystonia: clinical and therapeutic features in 85 patients. Arq Neuropsiquiatr 2008;66:15–21.ArticlePubMed
• 19. Hvizdošová L, Nevrlý M, Otruba P, Hluštík P, Kaňovský P, Zapletalová J. The prevalence of dystonic tremor and tremor associated with dystonia in patients with cervical dystonia. Sci Rep 2020;10:1436.ArticlePubMedPMC
• 20. Vu JP, Lee HY, Chen Q, Cisneros E, Barbano RL, Goetz CG, et al. Head tremor and pain in cervical dystonia. J Neurol 2021;268:1945–1950.ArticlePubMedPMCPDF
• 21. Godeiro-Junior C, Felicio AC, Aguiar PC, Borges V, Silva SM, Ferraz HB. Head tremor in patients with cervical dystonia: different outcome? Arq Neuropsiquiatr 2008;66:805–808.ArticlePubMed
• 22. Chen Q, Vu JP, Cisneros E, Benadof CN, Zhang Z, Barbano RL, et al. Postural directionality and head tremor in cervical dystonia. Tremor Other Hyperkinet Mov (N Y) 2020 Jan 20:10 [Epub]. https://doi.org/10.7916/tohm.v0.745. Article
• 23. Jankovic J, Adler CH, Charles D, Comella C, Stacy M, Schwartz M, et al. Primary results from the cervical dystonia patient registry for observation of onabotulinumtoxina efficacy (CD PROBE). J Neurol Sci 2015;349:84–93.ArticlePubMed
• 24. Trosch RM, Espay AJ, Truong D, Gil R, Singer C, LeWitt PA, et al. Multicenter observational study of abobotulinumtoxinA neurotoxin in cervical dystonia: The ANCHOR-CD registry. J Neurol Sci 2017;376:84–90.ArticlePubMed
• 25. Charles D, Brashear A, Hauser RA, Li HI, Boo LM, Brin MF, et al. Efficacy, tolerability, and immunogenicity of onabotulinumtoxina in a randomized, double-blind, placebo-controlled trial for cervical dystonia. Clin Neuropharmacol 2012;35:208–214.ArticlePubMed
• 26. Rodrigues FB, Duarte GS, Marques RE, Castelão M, Ferreira J, Sampaio C, et al. Botulinum toxin type A therapy for cervical dystonia. Cochrane Database Syst Rev 2020;11:CD003633.ArticlePubMedPMC
• 27. Jost WH, Schramm A, Müngersdorf M, Stenner A, Schwingenschuh P, Maisonobe P, et al. Effectiveness of botulinum neurotoxin type A injections in naïve and previously-treated patients suffering from Torti- or Laterocollis or -caput: results from a German-Austrian open-label prospective post-marketing surveillance study. J Neurol Sci 2019;399:44–50.ArticlePubMed
• 28. Papapetropoulos S, Baez S, Zitser J, Sengun C, Singer C. Retrocollis: classification, clinical phenotype, treatment outcomes and risk factors. Eur Neurol 2008;59:71–75.ArticlePubMedPDF
• 29. Hu Y, Pan L, Su J, Chen S, Zhang X, Pan Y, et al. Retroform cervical dystonia: target muscle selection and efficacy of botulinum toxin injection. Front Neurol 2022;13:952456.ArticlePubMedPMC
• 30. Esposito M, Peluso S, Dubbioso R, Allocca R, Iorillo F, Coppola A, et al. The occurrence of lateral shift in cervical dystonia. Neurol Sci 2017;38:683–686.ArticlePubMedPDF
• 31. Flowers JM, Hicklin LA, Marion MH. Anterior and posterior sagittal shift in cervical dystonia: a clinical and electromyographic study, including a new EMG approach of the longus colli muscle. Mov Disord 2011;26:2409–2414.ArticlePubMedPDF
• 32. Rudzińska M, Krawczyk M, Wójcik-Pędziwiatr M, Szczudlik A, Wasielewska A. Tremor associated with focal and segmental dystonia. Neurol Neurochir Pol 2013;47:223–231.ArticlePubMedPMC
• 33. Jost WH, Tatu L. Selection of muscles for botulinum toxin injections in cervical dystonia. Mov Disord Clin Pract 2015;2:224–226.ArticlePubMedPMCPDF
• 34. Su J, Hu Y, Djibo IM, Chen S, Pan Y, Zhang X, et al. Pivotal role of obliquus capitis inferior in torticaput revealed by single-photon emission computed tomography. J Neural Transm (Vienna) 2022;129:311–317.ArticlePubMedPDF
• 35. Hallgren RC, Rowan JJ. Implied evidence of the functional role of the rectus capitis posterior muscles. J Am Osteopath Assoc 2020;120:395–403.ArticlePubMed
• 36. Dressler D, Altavista MC, Altenmueller E, Bhidayasiri R, Bohlega S, Chana P, et al. Consensus guidelines for botulinum toxin therapy: general algorithms and dosing tables for dystonia and spasticity. J Neural Transm (Vienna) 2021;128:321–335.ArticlePubMedPMCPDF
• 37. Gerber TC, Gibbons RJ. Weighing the risks and benefits of cardiac imaging with ionizing radiation. JACC Cardiovasc Imaging 2010;3:528–535.ArticlePubMed
• 38. Tsao CW, Frost LE, Fanning K, Manning WJ, Hauser TH. Radiation dose in close proximity to patients after myocardial perfusion imaging: potential implications for hospital personnel and the public. J Am Coll Cardiol 2013;62:351–352.ArticlePubMed

Figure & Data

References

Citations

Citations to this article as recorded by  

Comments on this article