**Introduction to the Toxins Special Issue on Botulinum Neurotoxins in the Nervous System: Future Challenges for Novel Indications**

#### **Siro Luvisetto**

National Research Council of Italy-CNR, Institute of Biochemistry and Cell Biology (IBBC), via Ercole Ramarini 32, Monterotondo Scalo, 00015 Roma, Italy; siro.luvisetto@cnr.it

Received: 14 September 2020; Accepted: 14 September 2020; Published: 17 September 2020

*Botulinum* toxins (BoNTs) are a true wonder of nature. Like Dr. Jekyll and Mr. Hyde, they have a double "personality", making them unique among the toxins of bacterial origin. As Dr. Jekyll, BoNTs are drugs approved for a variety of clinical conditions while, as Mr. Hyde, they are one of the most dangerous toxins, causing botulism. In the past, many studies have extensively investigated the mechanism of action of BoNTs, showing a variety of apparently different mechanisms which have in common the block of the cholinergic transmission, mainly at the neuromuscular junction. These discoveries gave an extraordinary consensus to therapeutical use of BoNTs in human pathologies characterized by excessive muscle contractions, i.e., hypercholinergic dysfunctions including torticollis, blepharospasms, dystonia, and so on. Recently, the list of human disorders in which treatments with BoNTs have produced, or are expected to produce, beneficial effects is long and continuously growing. The ambitious goal of this Special Issue of *Toxins* was to provide an up-to-date picture on the state of studies for the development of new therapeutic treatments with BoNTs, mainly with serotypes A (BoNT/A) or B (BoNT/B). This Editorial is an introduction to the 25 contributions (14 research and 11 review papers) collected in this Special Issue of *Toxins*, which I strongly invite you to read in their original versions.

The first three papers focus on the treatment with BoNT/A of limb essential tremors (ETs), a neurology condition characterized by persistent postural, or kinetic, tremor due to involuntary rhythmic muscle activity of the upper or lower limbs, neck, and trunk. In detail, Zakin and Simpson [1] contributed with an overview on the techniques for BoNT/A injection, together with muscle targeting techniques, in the treatment of ETs. Niemann and Jankovic [2] reported the results of a retrospective study performed on a large database of patients treated with BoNT/A for hand tremor of different origins, mainly ETs but also dystonic, Parkinsonian, and cerebellar. Finally, Samotu et al. [3] analyzed the efficacy of BoNT/A in one open-label trial, with participants affected by Parkinson's disease (PD) and ETs.

Three clinical studies investigated the effects of BoNT/A in post-stroke spasticity (PSS), a common impairment arising from involuntary activation of muscles that often appears after stroke. Rosales et al. [4] performed a clinical trial (ONTIME) in post-stroke patients with spastic paresis, and analyzed the impact of BoNT/A on symptomatic spasticity progression. ONTIME provided evidence that an early BoNT/A injection improved muscle tone, delayed time to appearance of PSS symptoms, and significantly increased time until re-injection. Shin et al. [5] reported results from an open-label pilot study demonstrating that BoNT/A injection into finger and wrist flexors, followed by electrical stimulation of the finger extensor, improved active hand function in chronic stroke patients. Ianieri et al. [6] recalled the importance of performing an accurate evaluation of spasticity to determine how invalidating the symptoms are in order to personalize, for each patient, the optimal doses of BoNT/A, muscles, and injection time.

Another series of papers focuses on new dermatological uses of BoNTs. Kim et al. [7] contributed with a review for the use of BoNT/A in several off-label dermatological indications, including regenerative treatments of hypertrophic scarring and keloids, postoperative scar prevention, rosacea and facial flushing, and post-herpetic neuralgia, all conditions associated with hyperhidrosis, oily skin, psoriasis, and itching.

Itching constitutes another dermatological condition where application of BoNT/A may exert beneficial effects, especially in the treatment of neuropathic itching, a debilitating symptom appearing secondary to several skin, systemic, metabolic, and psychiatric disorders. The action of BoNT/A as an antipruritic agent is exhaustively reviewed by Gazerani [8] who summarizes all the evidence in favor both in animal models and in healthy human volunteers, and in many clinical conditions. The mechanism originating the antipruritic effects of BoNTs is discussed by Gazerani and also by Ramachandran et al. [9], who also give a possible explanation. The authors, analyzing the antipruritic effects of both BoNT/A and B in murine models, showed that both BoNT/A and B exert antipruritic effects in a mast cell/histamine-dependent and -independent manner.

Pain is another condition where the use of BoNTs is very promising. Different approaches have been adopted to treat chronic pain and, among others, the use of BOTOX® (commercial preparation of BoNT/A from Allergan, Inc., Irvine, CA, USA) has been recently authorized as a novel pharmacological indication for the prophylaxis of chronic migraine. In this context, Ion et al. [10] reported the results from a prospective study on the effect of a new BoNT/A formulation, namely XEOMIN® (commercial preparation of BoNT/A from Merz Pharmaceuticals, Inc., Frankfurt am Main, Germany), injected in patients with refractory chronic migraine. Unlike the other commercial preparationd of BoNT/A, XEOMIN® benefits from the absence of binding albumin protein, minimizing allergic reactions. The authors proved that XEOMIN® can be a prophylactic treatment in chronic migraine, effectively reducing the number of attack days, the number of migraine episodes per day, and the drug intake. Expanding the therapeutic uses of BoNTs, not only in pain, but also for overactive bladder, neurogenic detrusor overactivity, osteoarthritis, and wound healing, is exhaustively reviewed by Fonfria et al. [11]. The authors not only reviewed the effects of BoNTs, remote from injection sites, but also the effects of novel formulations, including modified and recombinant toxins, and of novel delivery methods, including transdermal, transurothelial, and transepithelial methods. Another potential analgesic effect of BoNTs is reviewed by Sandahl Michelsen et al. [12], who analyzed the results from a series of studies examining the effect of BoNT/A in alleviating pain in children with cerebral palsy (CP). The authors emphasize the difficulty concerning the treatment of pain in children with cerebral palsy (CP), a physical disability that affects the development of movement and posture in children and a neurological disorder in childhood caused by damage to either the fetal or the infant brain.

Continuing on the topic dedicated to the pain, Lee et al. [13] tested the efficacy of a lumbar sympathetic block with BoNT/A and B as pain therapy for complex regional pain syndrome (CRPS), a neuropathic pain syndrome causing spontaneous pain and allodynia. They found that the lumbar sympathetic block, with both BoNTs serotypes, constitutes a safe method to treat CRPS and, more surprisingly, BoNT/B is more effective and longer lasting than BoNT/A. The effects of BoNTs on CRPS as well as other neuropathic pain are also reviewed by Park et al. [14]. In addition to CRPS, this review also reports a complete overview of clinical studies of BoNT effects for central neuropathic pain, such as neuropathic pain after spinal cord injury, post-stroke shoulder pain, and central pain associated with multiple sclerosis.

In a basic science study, Finocchiaro et al. [15] compared the effect of BoNT/A and B in counteracting neuropathic pain in a murine model of sciatic nerve injury. The results confirmed that BoNT/A reduces neuropathic pain over a long period of time and, in parallel, it induces an acceleration of the regenerative processes of injured nerves, improving the functional recovery of the injured limb. BoNT/B can also reduce neuropathic pain over a long period of time, but, compared to BoNT/A, this reduction is not accompanied by an improvement in functional recovery. Finally, in an interesting review, Rojewska et al. [16] discussed whether and how BoNT/A reduces the development of neuropathic pain, with particular emphasis on spinal neuron–glia interactions and on the role of glial cells in BoNT/A-induced analgesia.

Another experimental study presented by Wang et al. [17] reported the nerve regeneration effects of BoNT/A on injured spinal cords in rats. What renders this paper unusual is the use of the BoNT/A heavy chain (BoNT/A-HC) as a catalytic subunit. This is completely in disagreement with the canonical view that BoNT/A-HC is the subunit necessary to bind to the vesicle presynaptic membrane and to

translocate, inside the vesicle, the BoNT/A light chain (BoNT/A-LC), which constitutes the catalytic subunit which, by cleaving the SNARE proteins, blocks the neuronal transmission. The authors found that local application of BoNT/A-HC to the site of spinal cord injury significantly induced an increased expression of growth-associated protein, together with a stimulation of neurite outgrowths. The mechanism by which BoNT/A-HC favors the relief of spinal motor dysfunction after nervous injury remains unknown.

As for novel indications of BoNTs, we should not forget that BoNTs have been considered as agents for inducing controlled paralysis in different muscles of the oral, maxillofacial, and temporomandibular joint region, with the aim to treat dysfunction and dislocation in clinical orthodontics and maxillofacial surgery. Clinical applications of BoNTs in treatment for the correction of severe malocclusion-associated problems, including occlusion after orthognathic surgery and mandible fracture, are reviewed by Seok and Kim [18]. This particular application of BoNTs is based on the principle that the induction of controlled paralysis of masticatory muscles reduces the tensional force to the mandible and prevents relapse, and affects maxillofacial bone growth and dental occlusion. Yoshida [19] performed a study comparing the treatment outcome after intramuscular injection of BoNTs in patients with recurrent temporomandibular joint dislocation (TMD). Restivo et al. [20] reported an interesting effect of BoNT/A in reducing hypersalivation in patients with neurological diseases of different etiologies, including Parkinson's, amyotrophic lateral sclerosis, brain injury, and cerebral palsy.

Two reviews focus on new therapeutic approaches of BoNTs in gynecology and urinary tract dysfunctions. In the first, Moga et al. [21] made a literature review regarding the efficiency of BoNT/A in the treatment of chronic pelvic pain, vaginismus, vulvodynia, and overactive bladder or urinary incontinence. In the second, Jhang and Kuo [22] did a literature review regarding treatment of neurogenic lower urinary tract dysfunction, such as overactive bladder, neurogenic detrusor overactivity, interstitial cystitis, urethral sphincter dyssynergia, dysfunctional voiding, benign prostate hyperplasia, and chronic prostatitis.

Moving on to completely different topics, Caleo and Restani [23] contributed with a review describing the experimental use of BoNTs as a tool to block synaptic function in specific brain areas, with central delivery of BoNTs used to treat pathological brain conditions such as epilepsy, cerebral ischemia, Parkinson's, and prion disease. For obvious reasons, primarily toxicity and toxin diffusion, these studies are still limited to animals. An example of this unusual utilization of BoNT/A is also reported by Antipova et al. [24], who injected toxin directly into the striatum of mice and compared the motor behavior. The authors speculate that locally applied BoNTs could be useful for treating brain dysfunctions that require the deactivation of local brain circuitry.

The last paper was contributed by Bano et al. [25]. This paper reports a relevant observation on the fact that a tetanus neurotoxin (TeNT), a relative of BoNT/B produced by a *Clostridia tetani* strain, is neutralized by antisera raised against BoNT/B. This finding implicates that, although TeNT is not considered a food-borne pathogen, it can be present in foodstuffs and interfere with the detection of *Clostridia botulinum* by the mouse test, giving rise to misleading results. It is interesting to recall that humans are not usually vaccinated against type B botulism but citizens in many countries are regularly vaccinated for tetanus. It might be interesting to investigate whether the human vaccine for type B botulism also protects from certain isoforms of TeNTs.

In conclusion, the papers included in this Special Issue of *Toxins* contributed to the advancement of the state of the art in the novel therapeutic uses of BoNTs. Furthermore, many of the published studies focused on emerging or less investigated applications of BoNTs, in particular for pathologies, thus providing the scientific community with new data supporting better knowledge of the contributions given by BoNTs to improve the health of humanity.

**Funding:** This research received no external funding.

**Acknowledgments:** As Guest Editor, I wish to thank all authors and colleagues who contributed to the success of this Special Issue of *Toxins*, and the expert peer reviewers, who performed careful and rigorous evaluations. The valuable contribution and editorial support of the MDPI management team and editorial staff are also acknowledged.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Botulinum Toxin in Management of Limb Tremor**

#### **Elina Zakin \* and David Simpson**

Department of Neuromuscular Medicine, Icahn School of Medicine at Mount Sinai, New York City, NY 10029, USA; david.simpson@mssm.edu

**\*** Correspondence: elina.zakin@mountsinai.org; Tel.: +19-212-241-6983

Academic Editor: Siro Luvisetto

Received: 24 October 2017; Accepted: 8 November 2017; Published: 10 November 2017

**Abstract:** Essential tremor is characterized by persistent, usually bilateral and symmetric, postural or kinetic activation of agonist and antagonist muscles involving either the distal or proximal upper extremity. Quality of life is often affected and one's ability to perform daily tasks becomes impaired. Oral therapies, including propranolol and primidone, can be effective in the management of essential tremor, although adverse effects can limit their use and about 50% of individuals lack response to oral pharmacotherapy. Locally administered botulinum toxin injection has become increasingly useful in the management of essential tremor. Targeting of select muscles with botulinum toxin is an area of active research, and muscle selection has important implications for toxin dosing and functional outcomes. The use of anatomical landmarks with palpation, EMG guidance, electrical stimulation, and ultrasound has been studied as a technique for muscle localization in toxin injection. Earlier studies implemented a standard protocol for the injection of (predominantly) wrist flexors and extensors using palpation and EMG guidance. Targeting of muscles by selection of specific activators of tremor (tailored to each patient) using kinematic analysis might allow for improvement in efficacy, including functional outcomes. It is this individualized muscle selection and toxin dosing (requiring injection within various sites of a single muscle) that has allowed for success in the management of tremors.

**Keywords:** botulinum toxin; limb tremors; muscle selection

#### **1. Introduction**

Essential tremor (ET) affects approximately 4–6% of individuals over the age of 65 [1]. Patients present with tremor characterized by persistent, bilateral, usually symmetric, postural, or kinetic tremor involving muscles of either the distal or proximal upper extremity. ET may also affect other body regions and functions including the voice, neck, lower limbs, and trunk. Over time, the severity of the tremor may worsen, and typically will affect daily tasks, including dressing, self-care, and feeding. Patients often present for evaluation and treatment once the tremor has begun to affect their quality of life. Oral therapies, including primidone, an anticonvulsant, and propranolol, a beta-adrenergic receptor antagonist, are effective in the treatment of essential tremor [2]. However, these agents reduce tremor amplitude by no more than 50%, and present a significant adverse side effect profile including dizziness, generalized fatigue, and bradycardia [3]. Additionally, about 30% of patients receive no therapeutic benefit from oral medications, leaving a relatively large population of individuals with pronounced ET untreated. Surgical options exist, including thalamotomy or implantation of either unilateral or bilateral thalamic deep brain stimulators, with Level C recommendation as 'possibly effective', but can only be performed in patients under the age of 75 [4]. Both pharmacologic and surgical options for the management of essential tremor, though effective for some, have their own potential for adverse events.

Local administration of intramuscular injections of botulinum toxin (BoNT) reduces excessive involuntary muscle activity and has emerged as an effective treatment for many movement disorders that are associated with muscle overactivity, including limb tremors. This article evaluates the current knowledge and evidence for the administration of botulinum toxin for limb tremors, including the process of muscle selection. This review will focus on techniques for botulinum toxin administration in limb tremors, as well as the safety/efficacy.

#### **2. Review of the Literature**

A tremor, which is an oscillatory movement produced by alternating or synchronous contractions of antagonistic muscles, is the most common movement disorder. Pharmacotherapy is usually not sufficient to control high-amplitude tremors, which can impair activities of daily living. Postural tremors respond more robustly to BoNT than do either kinetic or action tremors. In 1991, Jankovic, et al. reported the results of an open trial of BoNT in the treatment of 51 individuals with dystonic tremor, essential tremor, Parkinsonian tremor, peripherally induced tremor, and midbrain tremor [5]. They performed local injection for both head and limb tremor, noting that 67% of patients improved with an average latency from injection to response of 6.8 days. The average maximum duration of improvement was 10.5 weeks. This pilot study launched further investigation into the application of BoNT for selected movement disorders.

Shortly after this publication, Trosch and Pullman conducted an open-label study to determine the utility of BoNT injection in the management of severe hand tremors. They focused on forearm and arm tremors in 26 patients (12 with Parkinson disease, 14 with essential tremor), and used two clinical rating scales, subjective evaluations of function improvement and global disability, measures of weakness, and computer-assisted quantitative assessments of tremor to evaluate the effect of toxin injection after six weeks [6]. Although no change in clinical scores was noted, patients' disability scores (as measured by the Webster Tremor and Global Disability Scales) showed statistically significant differences between pre- and post-injection. Additionally, although amplitude differences were minimal, patients reported moderate to marked subjective improvement in functional benefit after injection. Thus, the study concluded that, while no major changes in clinical ratings or objective measures were noted, BoNT injections may subjectively improve tremor in some patients, particularly those with essential tremor.

In 1996, Jankovic, et al. performed the first randomized, double-blind, placebo-controlled study to evaluate BoNT-A in essential hand tremor. They studied 25 patients who were randomized to 50 units of botulinum type A compared to placebo. They evaluated rest, postural, and kinetic tremor at two- to four-week intervals over a 16-week period, using various tremor severity rating scales, accelerometry, and subjective improvement and disability scores [7]. They noted a significant improvement on the tremor severity scale at four weeks in the toxin group as compared to placebo control, with prolonged maintenance of the effect of toxin without significant effects on functional rating scales. Tremor evaluation using postural accelerometry showed a >30% reduction in amplitude in nine of 12 toxin-treated patients.

In 2000, Pacchetti et al. reported an open-label study of BoNT in 20 patients with disabling ET who did not respond to pharmacologic therapy, using activity of daily living self-questionnaires and tremor severity scales to establish patients' functional disability and tremor severity [8]. They noted a significant reduction in both severity and functional rating scales scores, as well as tremor amplitude reduction as measured with accelerometry and electromyography (EMG). They concluded that BoNT is safe and effective in reducing disability due to essential tremor.

In 2001, Brin et al. performed a multicenter, double-blind, placebo-controlled trial of botulinum toxin type A in essential hand tremor, studying 133 patients with ET, who were randomized to receive 50 or 100 units of botulinum toxin type A into both the wrist flexors and extensors, followed by a four-month follow-up [9]. The study showed significant improvement in postural tremor, with only minimal improvements in kinetic tremor and functional assessments.

#### **3. Process of Muscle Selection in Toxin Administration for Limb Tremors**

Early studies relied heavily on the clinical and electrophysiologic evaluation of patients' tremors to localize the muscles involved. In Jankovic's trial in 1991, careful clinical evaluation of the extremity was performed as it was held against gravity (to evaluate for postural tremor), during goal-directed movements (to evaluate kinetic tremor), or while performing specific activities such as writing (task-specific tremor). Surface electrode recording was performed on the limb in question, and muscle selection involved both agonist and antagonist groups (i.e., wrist extensors in addition to wrist flexors). Toxin was distributed into four to six different sites that were anatomically related to the muscles involved in the production of the tremor [5].

Trosch et al., working in 1994, also relied on clinical examination and EMG for tremor analysis. They used surface EMG of the flexor and extensor carpi radialis and ulnaris (FCU, FCR, ECR, and ECU), pronator teres, supinator, brachioradialis, flexor digitorum superficialis (FDS), extensor digitorum communis (EDC), biceps, and triceps muscles for each patient. Muscles that discharged rhythmically on needle EMG were injected, with almost every patient receiving wrist flexor and extensor injection [6]. Jankovic et al. used the clinical examination coupled with direct current amplification of EMG for the determination of accelerometry output for muscle selection in their randomized, double-blind, placebo-controlled study to evaluate toxin use in essential hand tremor [7]. Brin, et al. randomized patients to receive either 50 or 100 units, and used EMG guidance for BoNT administration [9]. Brin et al. and Jankovic et al. employed a fixed BoNT dosing schedule and a predetermined set of muscles. This is often currently not the case in practice, as dosages and selection of muscles injected are generally chosen individually based on tremor pattern. Pacchetti et al., in 2002, performed accelerometry and surface EMG to identify muscles with tremorogenic activity during impaired positions. They did not use EMG guidance for targeting muscles [8].

In 1997, O'Brien noted the importance of being proficient in electromyographic guidance and electrical stimulation for the improvement of the efficacy of therapy with BoNT injections [10]. More precise targeting of desired muscles for injection can be performed using these techniques, and assists the injector in preventing undesired adverse events. In using EMG guidance, the electromyographer's objective is to record motor unit potentials that are in close proximity to the needle tip, and, in so doing, to confirm the placement of toxin in the target tissue. Crisp, full-sized, bi- or triphasic motor unit potentials with fast rise times indicate that placement is near a contracting fascicle. This technique, coupled with clinical examination, was employed in early studies evaluating the efficacy of BoNT in the management of limb tremors.

#### **4. Techniques for Botulinum Toxin Administration for Limb Tremors**

More recent work on the use of botulinum toxin formulations has allowed for a different set of techniques in clinical decision-making and muscle selection with regards to toxin injection. Rahimi et al., in 2015, evaluated the use of sensor-based biomechanical patterns in assisting with incobotulinumtoxinA injection in upper extremity tremors of patients with Parkinson's disease [11]. They performed a 38-week open-label study on 528 patients with upper extremity tremor associated with Parkinson's disease, using kinematic technology to guide muscle selection in the injection of incobotulinumtoxinA. Specific kinematic measures of tremor amplitude allowed for detailed segmentation of tremor (with sensors placed at the fingers, hand, wrist, elbow, and shoulder) into components based on the direction at each joint in the upper extremity. Using specific MatLab® software, a detailed analysis of the directional contribution of each segment of the extremity was generated. Focusing on individual joints, a movement disorders neurologist used clinical judgment to determine the botulinum toxin dosage based on the kinematic measurement of tremor amplitude and the directional contribution of the tremor at each joint. This methodology tailored toxin injection to patient-specific tremor components. For the initial treatment, all participants were injected in the flexor carpi ulnaris and extensor carpi ulnaris muscles. At week 16, the Unified Parkinson's Disease Rating Scale (UPDRS) and Fahn–Tolosa–Marin (FTM) scale (measuring tremor severity) scores

showed improvement. The authors proposed that kinematics is a simple method for standardizing the assessment and treatment of upper extremity tremors in Parkinson's disease, which allows for optimal BoNT injection and more personalized tremor therapy for this select patient population.

In 2016 Samotus et al. used kinematically determined biomechanical patterns in a study of incobotulinumtoxinA injection of 24 patients with essential tremor, who were injected every four months [1]. Motor tracking devices were placed over the forearm, wrist, elbow, and shoulder joints, which captured tremor severity in angular root mean square (RMS) amplitudes and degrees of freedom. Dosage of toxin was determined based on the injector's interpretation of tremor severity and distribution of tremor, using kinematic recordings. Additional refinement of the technique was performed by comparing the change in tremor as measured kinematically between pre-injection to six weeks post- (peak effect of botulinum toxin) and 16 weeks post-treatment. Again, as in prior studies, the most frequently injected muscles were the FCR and ECR. This study was the first to use whole-limb kinematics to segment complex movements at multiple joints that were implicated in the tremor production. This helped target toxin injection more effectively and objectively, with targeted injections for each patient based on the unique tremor kinematics. Additional use of this kinematic technology to refine injection site selection and toxin dosage for subsequent injections was helpful for the improvement of essential tremor in these patients. The assessment of tremor movements in multiple degrees of freedom at the upper extremity joints was also useful in targeting toxin as it allowed for selection of the tremulous muscles, as opposed to only injecting the flexor and extensor wrist muscles with use of surface EMG electrodes, or a combination of surface EMG electrodes and electrical stimulation. This study concluded that the use of individualized injection parameters resulted in an improvement of tremor and disability in patients with essential tremor.

Jog and the Essential Tremor Study Team performed a prospective, randomized, double-blind, placebo-controlled multicenter study of 30 patients with essential tremor who underwent incobotulinumtoxinA versus placebo injection using TremorTrek® kinematic technology to evaluate tremor amplitude (and thus judge severity) and localize muscles of involvement. They concluded that kinematic analysis-based incobotulinumtoxinA therapy decreased tremor severity and improved hand function as compared to a placebo, with marked improvement in tolerability and reduced incidence of muscle weakness [12].

Because botulinum toxin works directly at the neuromuscular junction, it is important to obtain accurate placement reaching the end plate zone (usually located at the midpoint of the muscle fiber), which helps to increase clinical response by as much as 50% [13]. The dose of botulinum toxin is usually divided between several injection sites, depending on muscle size, with reports of up to eight injection sites per muscle. Use of anatomical localization via palpation and EMG guidance has been described, though additional techniques specific to the evaluation of tremor have been employed, as described above. Precision of muscle selection with EMG and electrical stimulation has allowed for a reduced amount of toxin to be used to produce the same benefit, which has also allowed for reduced frequency of neutralizing antibody formation. Incorporation of neuromuscular ultrasound to further target muscles of interest has allowed for even more precise localization in BoNT injection. While most experienced injectors believe that careful muscle targeting improves outcomes, the American Academy of Neurology (AAN) evidence-based guidelines published in 2008 concluded that there was insufficient Class 1 evidence to recommend any muscle targeting technique for limb injections of botulinum toxin [14]. The above guidelines were put forth using an extensive literature review with two class II studies evaluating botulinum toxin administration in upper extremity essential tremor. The class of evidence for the use of botulinum toxin for the treatment of tremor is rated as level C by the AAN. Controlled comparative trials are underway to further evaluate this issue.

#### **5. Safety/Efficacy of Toxin Use in Limb Tremors**

The side effects of BoNT are related in part to undesired diffusion of the drug from the muscle of interest to nearby muscles/structures. This can result in inadvertent weakness of muscles that were not targeted by the clinician. Patients should be educated on the risk of possible excess weakness, usually noted within the first few weeks after toxin injection.

Contraindications to treatment with BoNT include pregnancy, known impairment of neuromuscular transmission, and myopathy, as well as the presence of prior pareses [15]. The most common adverse effects are mainly injection-site- and dose-dependent. Systemic effects are infrequent and occur after administration of 10 times the dose typically used in practice. Additionally, the formation of neutralizing antibodies is possible; studies have reported the incidence to be 0.5% to 5% of patients treated with botulinum toxin (depending on toxin subtype) [16,17].

Factors associated with the formation of neutralizing Ab resistance include total BoNT dose administered and frequency of injection. This led to the accepted dogma and FDA labeling requiring a 12-week interval between injection sessions. Notably, most of these studies of resistance were conducted with an earlier formulation of ona-BoNT containing five times the amount of protein in current formulations. More recent data show a far lower incidence of neutralizing Ab formation. This issue has increasing importance given the potential advantage of the use of "booster injections" at shorter intervals, especially in limb tremor, where careful titration of low doses may be optimal. These issues are being evaluated in controlled prospective trials.

In many of the aforementioned trials, there was a high incidence of dose-dependent adverse events, including finger weakness, pain at injection sites, hematoma formation, and paresthesias. In many of the older studies, patients self-reported a high incidence of excessive weakness, which made true rater blinding difficult to achieve. This has improved with more focused muscle targeting using kinematics (as reported in the more recent literature). Additionally, focusing toxin injection on the flexor compartment (and avoiding injection of the extensor carpi muscles, unless they were the ones causing excessive disability) allowed for preservation of finger strength in more patients, further improving patient functional outcomes in performing daily activities such as writing, using a spoon, holding a cup, and pouring liquids.

In summary, BoNT injection should be considered for patients with limb ET who have not achieved relief with oral pharmacotherapy, or in whom the drug side effect profile is poorly tolerated. After careful evaluation of the tremor, experienced clinicians, using a specific injection pattern and muscle targeting techniques such as EMG, electrical stimulation, and ultrasound, can target the toxin to the muscles of interest. These targeting techniques (along with careful anatomic palpation) allow for more effective toxin delivery, and thus minimize the possibility of toxin diffusion and neutralizing antibody formation. Novel techniques, such as quantification of tremor components and amplitude using kinematic techniques, show promise in providing even more precise localization, improved efficacy, and reduced unwanted weakness. Though no controlled data exist, one can speculate that the combination of kinematic studies, muscle visualization, ultrasonography, and confirmation with electrical stimulation will allow for an even more targeted injection approach.

**Acknowledgments:** There are no sources of funding of the study to disclose.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Botulinum Toxin for the Treatment of Hand Tremor**

#### **Nicki Niemann and Joseph Jankovic \***

Parkinson's Disease Center and Movement Disorders Clinic, Department of Neurology, Baylor College of Medicine, Houston, TX 77030, USA; niemann@bcm.edu

**\*** Correspondence: josephj@bcm.edu; Tel.: +1-713-798-2273; Fax: +1-713-798-6808

Received: 26 June 2018; Accepted: 15 July 2018; Published: 19 July 2018

**Abstract:** The aim of this study is to review our longitudinal experience with onabotulinumtoxinA (onaBoNT-A) injections for medically refractory hand tremor. We performed a retrospective review of our database of patients treated with onaBoNT-A for hand tremor evaluated between 2010 and 2018 in at least 2 sessions with follow-up. The majority were injected into the forearm flexors (FF), although treatment was individualized. During the specified period, 91 patients (53 essential tremor, 31 dystonic tremor, 6 Parkinson's disease tremor, and 1 cerebellar outflow tremor) met our inclusion criteria. The mean age (SD) was 64.8 years (12.8), and mean duration of follow-up was 29.6 months (25.1) with mean of 7.7 (6.3) treatment visits. FF were injected in 89 (97.8%) patients, exclusively in 74 (81.3%), and 15 (16.5%) were injected in FF and other muscles. EMG guidance was used in 5 patients (5.5%). On a 0–4 "peak effect" rating scale (0 = no effect, 4 = marked improvement in severity and function), 80.2% and 85.7% of patients reported moderate or marked improvement (score 3 or 4) at their first and last follow-up visit, respectively. There was no statistically significant difference in the outcomes between first and last visit: average "peak effect" rating score (3.2 versus 3.4), "global" rating score (3.0 versus 3.2), latency of response (4.5 versus 3.8 days), and total duration of response (12.7 versus 12.8 weeks), except onaBoNT-A dose (65.0 versus 78.6 U/limb, *p* = 0.002). Of 1095 limb injections, there were 134 (12.2%) non-disabling and transient (mean 36 days) adverse events (132 limb weakness, 2 pain). OnaBoNT-A injections are safe and effective in the treatment of hand tremor.

**Keywords:** hand tremor; botulinum toxin; treatment; electromyography; kinematics; essential tremor; Parkinson's disease; dystonic tremor

**Key Contribution:** Botulinum toxin injections administered using surface anatomy are safe and associated with clinically meaningful and sustained improvement of hand tremor, regardless of etiology.

#### **1. Introduction**

Tremor is defined as an involuntary, rhythmic, oscillatory movement of a body part [1]. Hand tremor is a common movement disorder, often associated with impairment in quality of life [2]. Essential tremor (ET) is the most common cause of postural and kinetic hand tremor; 4.6% of individuals 65 years or older are thought to have ET [3]. Parkinson's disease (PD) affects 0.43–1.90% of individuals 60 years or older [4] with the majority experiencing rest or postural tremor [5,6]. While rest tremor is typically associated with PD, most PD patients also have a postural or re-emergent tremor, which is the most disabling form of PD-related tremor [6,7]. The prevalence of dystonic tremor is unknown [8], although hand tremor has been reported in up to 70% of patients in a series of patients with dystonia of various types [9]. There are many other causes of tremor besides ET, PD, and dystonia, which can affect not only hands but also other body parts [1]. Treatment of tremor with oral medications often provides insufficient relief, especially when the tremor is severe, and is frequently associated

with systemic side effects [10,11]. Deep brain stimulation (DBS) is a highly effective treatment for tremor in ET and PD, but is invasive and can be associated with stimulation-induced and long-term side effects [12,13]. Functional lesional neurosurgery, especially focused ultrasound, has garnered more interest recently in the treatment of tremor but this intervention is not readily available and is associated with potentially serious adverse effects, particularly when performed bilaterally [14–16].

Botulinum toxin (BoNT) has been increasingly used in the treatment of focal tremors with reported good outcomes and long-term safety [17,18]. One of the main advantages of this treatment over above-noted strategies is the lack of systemic side effects. The aim of the present study is to describe our long-term experience with onabotulinumtoxinA (onaBoNT-A) in the treatment of medically refractory hand tremor using an individualized approach.

#### **2. Results**

We identified a total of 91 patients who were treated with onaBoNT-A for hand tremor at least twice and had adequate follow-up during the period from 1st January 2010 to 1st January 2018. Of these patients, 23 (25.3%) received their initial injection prior to 1st January 2010. Diagnoses included ET (53 patients with 395 visits), PD (6 patients with 27 visits), dystonia (31 patients with 269 visits), and cerebellar outflow tremor (1 patient with 13 visits) (Figure 1). The male to female ratio was 1.1:1 (48 male and 43 female) and the mean age of all participants was 64.8 years at the time of their initial injection visit (Table 1). The mean duration of symptoms prior to the first onaBoNT-A injection was 23.2 years. The mean duration of follow-up from the first to the last injection visit was 29.6 months during which time patients underwent a mean of 7.7 injections.

**Figure 1.** Etiology of hand tremor in patients treated with onabotulinumtoxinA (onaBoNT-A) injections from 1st January 2010 to 1st January 2018.



SD = Standard deviation, ET = Essential tremor, PD = Parkinson's disease, onaBoNT-A = onabotulinumtoxin-A, COT = Cerebellar outflow tremor. \* *p* = 0.49 for comparison with ET. † mean dose per limb (first and last visit combined).

Forearm flexors were injected in 89 (97.8%) patients, 74 (81.3%) received injections exclusively to the forearm flexors, and 15 (16.5%) received injections to the forearm flexors and at least one other upper extremity muscle compartment (hand muscles, forearm extensors, arm flexors, arm extensors, and/or shoulder abductors); 2 (2.2%) received no forearm flexor injections. The specific muscles injected according to tremor etiology are listed in Table 2. EMG was used in 5 (5.5%) patients while the rest were injected using surface anatomy. The mean dose of onaBoNT-A increased significantly from 65.0 to 78.6 (*p* = 0.002) between the first and last visit; the mean dose was 71.8 U per limb. There was no statistically significant difference between outcomes from the first and last visit with respect to all other outcome measures: peak effect score was 3.2 versus 3.4 (*p* = 0.17), global rating was 3.0 versus 3.2 (*p* = 0.06), and latency of response was 4.5 versus 3.8 days (*p* = 0.10) (Table 3). Using the peak effect rating scale, 80.2% and 85.7% of patients rated their improvement as either moderate or marked (score of 3 or 4) at their first and last visit, compared to 74.7% and 80.2% of patients on the global effect rating scale at the first and last visit, respectively. A separate analysis comparing outcomes on the global rating and peak effect rating scale within each group (e.g., ET outcomes after first versus last injection) and between groups (e.g., ET versus dystonia) did not reveal any statistically significant differences (data not shown). The mean duration of maximum response at the first and last injection was 12.1 and 12.7 weeks (*p* = 0.70), respectively. The mean total duration of response at the first and last injection was 12.7 and 12.8 weeks (*p* = 0.87), respectively, but this is likely an underestimate as the effects of the prior injection had not fully worn off in 27 (29.7%) and 41 (45.1%) patients at the time of follow-up after the first and last visit. Of the total 1095 limbs injected, 134 (12.2%) were associated with some adverse effect (120 hand grip weakness, 1 focal finger flexor weakness, 11 elbow flexor weakness, and 2 pain), none of which were considered to be severe or disabling. The mean duration of side effects was 36 days (range 7–120).


**Table 2.** Injection strategy at the last visit according to tremor etiology (*n* = 91).

ET = Essential tremor, PD = Parkinson's disease, COT = Cerebellar outflow tremor, FCU = Flexor carpi ulnaris, FCR = Flexor carpi radialis, FDS = Flexor digitorum superficialis, ADM = Abductor digiti minimi, APB = Abductor pollicis brevis, ED = extensor digitorum, EPB = Extensor pollicis brevis, UNS extensor = unspecified extensor muscle.

**Table 3.** Outcomes following onaBoNT-A injections for hand tremor (*n* = 91).


OnaBoNT-A = onabotulinumtoxin-A. \* *p* < 0.05 indicating a statistically significant result. † Data not included for patients with persistent response (27 versus 41) or lack of data (18 versus 13) at the time of follow-up after the first and last injection.

At the end of the study period (1st January 2018), 41 (45.1%) patients were still receiving onaBoNT-A injections. The remaining 50 (54.9%) patients were no longer receiving injections in our clinic: 5 (5.5%) were lost to follow-up, 14 (15.4%) were dissatisfied and discontinued treatment usually after only 2–3 injections, and 4 (4.4%) transitioned to alternative therapy; the reason for discontinuation was financial or unknown in 24 (26.4%) and 3 (3.3%) died of unrelated causes. Additional 30 patients (ET, *n* = 14; dystonia, *n* = 10; other, *n* = 6) did not meet our inclusion criteria as they discontinued onaBoNT-A injections after just 1 visit, either for unknown reasons (*n* = 15), lack of benefit (*n* = 10), adverse effects (*n* = 2), or other reasons (*n* = 3).

#### **3. Discussion**

In this study we provide longitudinal, follow-up, data for 91 patients treated with onaBoNT-A for hand tremor over an average period of 2.5 years. To our knowledge, this is the longest duration of follow-up data reported regarding the use of BoNT for hand tremor, although we previously reported our long-term experience with BoNT in the treatment of dystonia over a period of more than 20 years [19]. The mean age of our patient population and the male predominance is similar to those reported in recently published studies of BoNT in the treatment of tremor [20–25]. There was a relatively long period of time between onset of symptoms and first injection (mean 23.2 years, up to 75 years). In contrast to our experience with BoNT for dystonia [19], we found no significant difference in the outcome measures (peak effect, global effect, latency of response, or duration of response) between the first and last visit. We found no difference in response to BoNT injections based on underlying

etiology of the tremor as measured by the peak effect and global rating. There was, however, a modest but significant increase in the average dose of onaBoNT-A, from 65.0 U to 78.6 U (difference of 13.6 U; *p* = 0.002). The most plausible reason for the 21% increase in average BoNT dose per limb between the first and last visit is that the initial dose is often quite conservative and relatively lower than the estimated maintenance dose to minimize potential adverse effects such as hand weakness. We have observed sustained benefit during the follow-up period (mean, 29.6 months; range, 3–88 months) with a significant proportion of patients reporting moderate or marked improvement of tremor severity and function (score of 3 or 4) on the peak effect rating scale (80.2% versus 85.7%) and global effect rating scale (74.7% versus 80.2%) at the first and last visit. Of the 50 (54.9%) patients who had discontinued treatment with BoNT injections during the study period, 32 (35.2%) did so for unknown or unavoidable reasons, while only 14 (15.4%) were dissatisfied with treatment.

In the earliest study of onaBoNT-A in the treatment of tremor from our center, 10 patients with hand tremor (mixed etiology) received a mean dose of 95 ± 38 U predominantly divided between the forearm flexor and forearm extensor compartment [26]. The peak effect hand tremor was lower (2.0 ± 1.7 versus 3.2 ± 1.3) and the rate of focal weakness was higher (40.0% versus 12.1%) compared to the data reported in the present study. This is consistent with data from the first two double-blind, placebo-controlled trials of BoNT in the treatment of ET [27,28]. In both studies, a pre-determined set of muscles (forearm flexors and extensors) was injected with a fixed dose of BoNT, yielding clinically significant improvement of tremor, however with dose-dependent hand and finger weakness in up to 92% of patients. Because of the high frequency of extensor finger weakness we have since changed our injection technique to favor the forearm flexors while largely avoiding the forearm extensors, thereby reducing the frequency of clinically significant (extensor) wrist and finger weakness while maintaining tremor reduction (Figure 2) [29].

Several well-designed studies have evaluated the use of kinematic tremor analysis coupled with EMG-guidance in the application of BoNT [20,21,23,30–32]. Kinematic analysis allows the clinician to objectively characterize tremor at the wrist, elbow and shoulder joint, thereby creating an individual "tremor profile" which can be re-assessed at future visits thus allowing for optimization of the injection pattern. In a study involving 30 patients with ET who were randomized to receive a single injection of either incoBoNT-A (*n* = 19) or placebo (*n* = 11) using a kinematic analysis-based approach the mean total dose of incoBoNT-A was 116.3 ± 53.0 U distributed between multiple muscle compartments [23]. There was a significant reduction in wrist tremor amplitude and in the Fahn-Tolosa-Marin (FTM) tremor rating scale part B at week 4 and 8 post-injection. The injections were associated with approximately 15–20% reduced grip strength in the treatment group at week 4 and finger extensor weakness in 2 (10.5%) patients. There was no long-term data provided in this study but the same investigators conducted an open-label, long-term (96 weeks) study of kinematic analysis-guided incoBoNT-A injections for ET (*n* = 24) and PD (*n* = 28) upper limb tremor [32]. In this study, a mean of 9.4 ± 2.3 versus 10.2 ± 2.6 muscles were injected with a mean of 180.3 ± 74.8 versus 188.5 ± 78.1 U incoBoNT-A in PD and ET, respectively, at the time of the last injection visit. There was a statistically significant reduction of rest and action tremor on items 20 and 21 of the Unified Parkinson's Disease Rating Scale (UPDRS), FTM tremor rating scale part A–C, and tremor amplitude. A total of 6 patients (11.5% [4 PD and 2 ET]) withdrew from the study due to disabling weakness and another 5 patients (9.6% [3 PD and 2 ET]) withdrew due to lack of benefit. While the rate of significant weakness fluctuated throughout the study, 5.0–29.4% of PD and ET patients experienced significant finger extensor weakness after each injection session.

In a double-blind, placebo-controlled, cross-over study of 30 patients with PD-related rest and action tremor, Mittal and colleagues injected incoBoNT-A with EMG guidance to identify muscles with tremor activity [22]. IncoBoNT-A or placebo (saline) was injected at the start of the trial with cross-over to the opposite treatment arm at week 12; outcome was assessed at week 4 and 8 after each injection. A mean of 9 (range 7–12) injections into forearm and hand muscles with a mean of 100 U (range, 85–110) incoBoNT-A per patient per visit were performed. IncoBoNT-A injections were associated with significant reduction of tremor affecting activities of daily living as well as rest and action tremor (UPDRS items 16, 20, and 21) with low rates of non-disabling (10%) and disabling (6.6%) hand weakness. Using a similar study design, the same authors performed incoBoNT-A injections in 28 patients (of 33 enrolled) with hand ET [25]. IncoBoNT-A were injected into a mean of 9 (range, 8–14) hand and forearm muscles (80–120 U). IncoBoNT-A treatment led to significant reduction of tremor (FTM tremor rating scale part B and the National Institute of Health Collaborative Genetic Criteria tremor score severity) at weeks 4 and 8 (*p* < 0.05). IncoBoNT-A treatments were safe and well-tolerated, although mild hand weakness occurred in 21% and 1 patient withdrew from the study due to disabling hand weakness.

Although EMG, ultrasound, and kinematic-guided analysis can be utilized to successfully treat hand tremor with BoNT, no muscle targeting technique has yet been proven to be superior [33,34]. The results described in our study, using palpation and surface anatomy, seem similar to those obtained using kinematic analysis and/or EMG-guided injections [24,27,32,35]. Thus, it is unclear if a technology-guided approach is necessary and whether it yields better outcomes. Indeed, 21.2% of 52 patients treated using kinematic analysis and EMG-guided injections by Samotus et al. [32] withdrew because of adverse effects or lack of benefit, compared to only 15.4% of our 91 patients, although the two populations are not comparable. Technology-guided injections are useful in some instances, such as when the target muscles are difficult to identify, however, it is nearly impossible to sample all 20 forearm muscles plus other, more proximal, muscles to identify all the muscles that contribute to the tremor and target them for BoNT. Even when EMG is used to target the identified muscle the precise site of the tip of the electrode is impossible to validate with any certainty (even with ultrasound) and it is likely that the effects of BoNT will diffuse beyond the boundaries of the intended target [17]. Furthermore, EMG may sometimes be misleading, particularly in patients injected for dystonia, as it may not differentiate between the primary (agonist) and the compensatory (antagonist) muscle contraction, potentially resulting in injection of the wrong muscle. In this regard, evaluation for mirror movements may be helpful [36], particularly in patients with dystonic tremor. Finally, and perhaps most importantly, technology-guided injections are more time-consuming, more painful and more costly [37] without meaningfully improving the outcomes [38].

In our longitudinal study we demonstrated that a comparatively low dose of onaBoNT-A injected into a very limited number of muscles (forearm flexors were injected exclusively in 81.3% of patients) guided by palpation and surface anatomy produced clinically meaningful and sustained improvement of tremor in the majority of patients with low rates of non-disabling focal weakness (12.1%). Although we mainly injected the forearm flexors, the injection pattern should always be individualized and optimized based on the characteristics of the patient's tremor. For an instance, injection of the biceps brachii or pronator teres muscle may be considered in the treatment of the classical supination/pronation tremor in PD [39].

The main strength of our report is that it provides a "real world" account of long-term experience with the use of BoNT in hand tremor using a pragmatic, clinical approach. However, we acknowledge that our study has some limitations, particularly due to its open-label and retrospective design. Further, of 30 patients who discontinued treatment with BoNT after only 1 session, 12 (40%) did so due to either troublesome side effects or lack of benefit. We would not, however, consider these as "failures" since 2 or more treatment visits are often required to optimize the outcome. We do not think that bias and placebo response account for the reported outcomes as patients had sustained improvement over time, noticed wearing off between doses, and returned repeatedly for injections, many traveling long distances. Furthermore, it is important to point out that nearly all our patients were previously treated with optimal medical therapies but because of poor response or undesirable side effects were selected for treatment with BoNT. Prospective studies comparing different methods of muscle selection, localization, and injection techniques may provide further validation of our results.

#### **4. Conclusions**

In conclusion, onaBoNT-A injections are safe and associated with clinically meaningful and sustained improvement of hand tremor in a mixed population of patients with medically refractory hand tremor.

#### **5. Materials and Methods**

#### *5.1. Study Description*

We conducted a retrospective analysis of our medical records of patients treated with onaBoNT-A injections for hand tremor in the Parkinson's Disease Center and Movement Disorders Clinic (PDCMDC) at Baylor College of Medicine (BCM). The study was approved by the BCM Institutional Review Board (IRB), number H-42714, approved January 31st, 2018.

#### *5.2. Data Collection*

Patients treated with BoNT for tremor were identified through review of the electronic medical record (EMR) at BCM. The medical history, past treatments, examination and the indication for BoNT treatment was determined by review of the medical record. The "Botulinum Toxin Data Form" ("BoNT Data Form"), which captures injection data (muscle selection, dose, type of BoNT, and response to prior injection) was completed at the initial and all subsequent injection visits. The "peak effect", one of the two main outcome measures, was defined as the maximum improvement of tremor following BoNT injection, and was assessed by a 0–4 rating scale (0 = no effect; 1 = mild effect, but no functional improvement; 2 = moderate improvement, but no change in functional disability; 3 = moderate change in both severity and function; and 4 = marked improvement in severity and function). The "global rating", the other outcome measure, was defined as the peak effect score minus 1 point for a mild side effect, such as transient pain or mild weakness, and minus 2 points for a disabling side effect, such as marked weakness causing impairment of function. The "latency of response" was the time from injection of BoNT until the first noticeable response (in days), "duration of maximum response" was the period of time during which the patient experienced peak effect (in weeks), and "total duration of response" represented the time from the first noticeable response to complete wearing off (in weeks). The type, severity, and duration (in days) of side effects were also recorded.

#### *5.3. Selection Criteria*

Inclusion criteria for this study were: (1). Presence of troublesome hand tremor that is refractory to medical therapy, including optimized trial of propranolol, primidone, topiramate for ET; levodopa, dopamine agonists or anticholinergics for PD tremor; and anticholinergics for dystonic tremor, (2) chemodenervation using onaBoNT-A injections of the arm(s) for hand tremor, (3) minimum of two BoNT injection visits between 1st January 2010 (introduction of EMR at BCM), and 1st January 2018 with follow-up data, (4) follow-up within 1 year after the first and last injection, and (5) age older than 18 years at the time of the first injection. Patients first treated prior to introduction of the EMR were included in this study, but their "initial visit" was recorded as their first BoNT injection visit after introduction of the EMR. Injection and outcome data were analyzed for the first and last visit while information regarding side effects (type and duration) were captured for all visits. This approach was chosen for simplicity since some patients had more than a dozen injection visits during the study period which would make it difficult to analyze all outcomes at each visit separately. We further recorded the reason for discontinuation of onaBoNT-A injections (if known).

#### *5.4. Botulinum Toxin Injection*

We individualized the target muscle and dosage based on the amplitude and character of the tremor, starting at a relatively low dose and usually optimizing the treatment over subsequent 2–3 visits. The majority of patients received an initial injection of approximately 25–75 units (U) of onaBoNT-A per limb divided between flexor carpi radialis (2/3) and flexor carpi ulnaris (1/3) (Figure 2). We do not routinely utilize electromyography (EMG) or ultrasound, but rely on palpation and surface anatomy for guidance in most BoNT injections.

**Figure 2.** Localization of the forearm flexors most commonly injected in hand tremor. Reprinted with permission from [29], Copyright Cambridge University Press, 2003.

#### *5.5. Statistical Analysis*

Using descriptive statistics, we analyzed the following information: age at first injection visit, disease duration, number of injection visits, duration of treatment with onaBoNT-A, average dose of onaBoNT-A per limb, peak effect, global rating, latency of response, duration of maximum response, and total duration of response. A paired *t*-test was used to compare continuous data outcomes (e.g., BoNT dose) at the first and last injection visit. Ordinal outcomes (e.g., peak effect) were compared using nonparametric tests.

**Author Contributions:** Conceptualization, J.J.; Methodology, N.N., J.J.; Validation, N.N., J.J.; Formal Analysis, N.N.; Investigation, N.N.; Resources, J.J.; Data Curation, N.N.; Writing-Original Draft Preparation, N.N.; Writing-Review & Editing, N.N., J.J.; Visualization, N.N.; Supervision, J.J.; Project Administration, N.N., J.J.

**Funding:** This research received no external funding.

**Conflicts of Interest:** Nicki Niemann: The author declare no conflict of interest. Joseph Jankovic: Dr. Jankovic has received research and/or training grants from: Adamas Pharmaceuticals, Inc.; Allergan, Inc.; Biotie Therapies; CHDI Foundation; Civitas/Acorda Therapeutics; Dystonia Coalition; Dystonia Medical Research Foundation; F. Hoffmann-La Roche Ltd.; Huntington Study Group; Kyowa Haako Kirin Pharma, Inc.; Medtronic Neuromodulation; Merz Pharmaceuticals; Michael J. Fox Foundation for Parkinson Research; National Institutes of Health; Neurocrine Biosciences; NeuroDerm Ltd.; Parkinson's Foundation; Nuvelution; Parkinson Study Group; Pfizer Inc.; Prothena Biosciences Inc.; Psyadon Pharmaceuticals, Inc.; Revance Therapeutics, Inc.; Sangamo BioSciences, Inc.; St. Jude Medical; Teva Pharmaceutical Industries Ltd. Dr. Jankovic has served as a consultant or as an advisory committee member for: Adamas Pharmaceuticals, Inc.; Allergan, Inc.; Merz Pharmaceuticals; Pfizer Inc.; Prothena Biosciences; Revance Therapeutics, Inc.; Teva Pharmaceutical Industries Ltd. Dr. Jankovic has received royalties or other payments from: Cambridge; Elsevier; Future Science Group; Hodder Arnold; Medlink: Neurology; Lippincott Williams and Wilkins; Wiley-Blackwell. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
