Long-Term Outcomes of Deep Brain Stimulation for Pediatric Dystonia Free

Dystonia is a movement disorder defined by involuntary sustained or intermittent muscle contractions, typically described as patterned and twisting postures. They are commonly action-induced, may be associated with tremor or overflow muscle activation, and are alleviated by a “sensory trick.” Pediatric dystonia is defined by the onset at less than 21 years of age, with a prevalence estimated at 2–50 cases per million [1].

Classification of dystonia distinguishes (1) inherited dystonia, with proven genetic origin, including both isolated dystonias and those associated with heredodegenerative disorders; (2) acquired dystonia, which can be referred to a known cause, including perinatal brain injury, infection, toxins, and vascular events; and (3) idiopathic dystonia, when the cause is unknown [2]. Dystonia can also be described by age of onset (early or late), body distribution (generalized, segmental, multifocal, or focal), temporal pattern, and associated features. Pediatric dystonias are typically generalized, with DYT1 mutations accounting for 80–90% of cases in Ashkenazi Jewish populations and 16–53% of cases in non-Jewish populations. Other pediatric dystonias can be attributed to one of the other known 27 DYT mutations, as well as other heredodegenerative diseases and acquired causes [1].

There are both oral and injectable options for medical therapy, although their efficacy is limited [3]. As botulinum toxin injections are not a feasible option in most pediatric cases with generalized dystonia, surgical therapy assumes a prominent role in cases refractory to medications or when patients are unable to tolerate medications due to side effects. Among surgical treatments of dystonia, deep brain stimulation (DBS) has emerged as a relatively safe and effective procedure. Although the mechanism of action is not entirely clear, it is thought that chronic stimulation disrupts the abnormal function of thalamic and basal ganglia nuclei in dystonia. A major benefit of DBS is that it is reversible and can be used bilaterally without significant risk of cognitive or motor dysfunction, as previously seen in cases treated with pallidotomy [4]. Ideal candidates for DBS have generalized, segmental, or multifocal dystonia that is refractory to medications and are experiencing significant disability due to impairment of motor function or pain affecting quality of life [5]. Shorter duration of disease, young age, and specific genetic conditions (DYT1) are considered predictors of better outcome [6‒8]; thus, genetic testing may be helpful prior to surgery. Factors that may make DBS a less desirable option include age younger than 7 years, low weight, high risk for infection, poor follow-up, and high risk of suicide [9].

Initial case reports and a series of pallidal DBS for dystonia dating back to the early 2000s demonstrated extremely encouraging results, particularly in primary generalized dystonias [10‒13]. Blinded studies comparing pallidal DBS to sham stimulation in isolated generalized and segmental dystonia showed Burke-Fahn-Marsden dystonia rating score (BFMDRS) improvements ranging from 39.3% at 3 months to 54.6% at 12 months [14, 15]. Additional studies have confirmed remarkable short-term benefits in children, albeit with some variability depending on the underlying etiology, in a wide variety of dystonia subtypes (Table 1) [8, 16‒35].

In addition to the accuracy of surgical placement for each individual patient (see later), the ability of the neurologist/programmer to set and adjust multiple stimulation parameters through DBS programming is crucial to optimal clinical outcomes [36]. Most clinicians prefer activating one or two ventral contacts with monopolar stimulation [36], a choice supported by multiple clinical series in adults as well as research based on calculation of effective volumes of tissue activation. These have shown that stimulation of the ventroposterior globus pallidus internus (GPi) is associated with the best motor outcomes [37‒40]. While initial case series have reported positive responses with higher pulse widths up to 450 μs and frequencies of 130 Hz and higher [13], these parameters can rapidly deplete the internal pulse generator (IPG) battery. Later studies have shown equal efficacy with lower frequencies [41, 42] and narrower pulse widths [43].

Despite the undisputable efficacy of DBS in the short term, there is limited literature supporting the long-term benefits in the pediatric population. Some studies have shown a persistent [38, 44] or even improved effect [45], but others have suggested the potential for dystonia recurrence [46, 47]. In addition, there is concern about the long-term tolerability of DBS implants. Further analysis is therefore warranted, as pediatric patients will likely live with DBS implants for multiple decades or their entire life. They, and their parents, need to be counseled appropriately on both the expected long-term benefits and possible adverse effects, including hardware complications or neuropsychiatric impact.

Over two decades of experience have codified DBS surgery procedures, with some variations between adult and pediatric patients. Stereotactic surgery for DBS device implantation is typically performed in two parts: first, the stimulating lead is placed into the target of choice, and second, the extension cable and pulse generator are implanted. While a single-stage procedure may be more efficient, it is associated with a higher rate of surgical infection [48]. The target is identified either by estimate coordinates from brain atlases or with direct visualization on imaging, with particular attention to trajectory to avoid lateral ventricles and cortical veins to decrease the risk of hemorrhage [49]. Children with severe dystonia, frequently complicated by developmental delay, are often unable to tolerate awake surgery due to the difficulty cooperating with instructions and risk of involuntary movements while in a stereotactic frame [50, 51]. Therefore, different than adult practices, most DBS implants in pediatric patients are performed under general anesthesia [52]. Alternatively, a combination of dexmedetomidine and propofol or remifentanil and ketamine can be safely used to obtain adequate sedation and analgesia without respiratory depression [53], preserving microelectrode recordings and awake cooperation with intraoperative testing [50]. In those cases where asleep placement is preferred, frame-based stereotactic surgery is performed without microelectrode recordings and confirmed postoperatively with either computed tomography (CT) or magnetic resonance imaging [54, 55]. The downside to this approach is that accuracy becomes dependent uniquely on the frame system [56], in addition to the logistical difficulty of transporting patients between different locations. Alternatively, intraoperative CT images can be obtained and then fused to preoperative MRI to confirm lead placement [57‒59], a procedure possibly limited by the accuracy of the image fusion software [58, 60]. The use of interventional MRI-guided DBS using a skull-mounted aiming device in conjunction with dedicated software, which is independent of physiological testing, showed good accuracy and comparable outcomes in pediatric dystonia [29, 61]. Robot-assisted stereotactic implantation may have even greater accuracy [62].

Optimal targets may vary, particularly for children with acquired dystonia who may have structural lesions within the basal ganglia making standard GPi placement more difficult to visualize or less responsive to DBS [63]. There is no data directly comparing GPi versus subthalamic nucleus (STN) or thalamic stimulation in the pediatric population. STN stimulation demonstrated good outcomes in one pediatric study of 9 patients with isolated dystonia [35]. A phase 1 trial exploring bilateral thalamic DBS targeting the nucleus ventralis intermedius/ventral oral posterior (Vim/Vop) in pediatric patients with acquired dystonia showed outcomes comparable to pallidal stimulation [64]. Temporary implantation of depth electrodes can provide neurophysiological testing of potential target sites prior to permanent insertion [51].

About 1–4 weeks after brain electrode implantation, a second surgery is performed to place the IPG, typically in the chest below the clavicle bone, like a cardiac pacemaker. Implanted brain leads are connected to an extension wire tunneled behind the ear and down the neck to reach the IPG in the chest. About 1 week after IPG placement, to allow for wound healing, a clinician can proceed to perform initial programming using a handheld device to communicate with the IPG and achieve therapeutic stimulation.

Several innovations have advanced the DBS field over the past few years. Directional leads have been created to improve current steering, using segmented contacts designed to prevent current spread to nearby structures, possibly associated with side effects [65]. There is limited data on the use of directional leads in dystonia, but studies exploring the use of computational VTA models have demonstrated potential benefit [66]. Sensing technologies able to measure local field potential (LFP) recordings are also being studied. LFP theta band activity within the GPi seems to correspond to dystonia severity [67] and is maximal at optimal contacts [68]. In the future, clinicians may be able to use this information to readily identify the best DBS settings or create a closed-loop or adaptive neuromodulatory algorithm. New visualization tools provide 3D reconstructions of lead placement using patient-specific anatomical plates derived by MRI and CT imaging [69]. Anatomical visualization will likely simplify programming by guiding the choice of active electrodes and size of the electrical field, in addition to help with troubleshooting when lack of benefit or intolerable side effects ensue. Lastly, technology that allows remote DBS programming via telemedicine was recently approved by the FDA [69], a development that will improve access to DBS treatment in rural areas.

We conducted a literature search using PubMed for case reports, case series, reviews, or clinical trials prior to April 1, 2021, reporting data on pediatric dystonia patients who underwent DBS implantation (age <21 years at surgery). We used the following search terms were used in combination to identify potential articles: “pediatric” OR “children,” and “dystonia,” and “deep brain stimulation.” Abstracts and full-text articles were reviewed. The references of each article were further reviewed for additional studies that were not captured by the original search. A total of 217 studies were retrieved. Additional search phrase “long term” was used to capture articles that discussed long-term motor, neuropsychiatric, and/or quality of life outcomes, as well as adverse events. This yielded 44 articles, 20 of which included individual patient data demonstrating at least one patient who had post-DBS follow-up of at least 5 years and BFMDRS motor scores at last follow-up. Five articles discussed long-term outcomes after DBS placement for status dystonicus.

Patients with inherited dystonia without central nervous system pathology have been historically preferred candidates for DBS due to the possible expectation of complete symptom resolution in the absence of an underlying structural lesion or degeneration. The most common gene mutation found in childhood-onset generalized dystonia is DYT1, which is the result of a GAG deletion in the gene encoding torsin A. It is inherited in an autosomal dominant pattern with 30–40% penetrance as well as variable phenotypic expression [70]. Multiple studies have shown a robust response to GPi DBS in pediatric patients with DYT1, ranging from 40.5%–88.3% as measured by mean BFMDRS motor score and 50%–76.8% as measured by BFMDRS disability score [7, 18‒20, 26, 71‒74].

DYT6 dystonia is caused by mutations of the THAP1 gene and is inherited in an autosomal dominant fashion with about 60% penetrance. Symptom onset is in the cranio-cervical region in about half of patients, and in the upper extremities in other patients [75]. While experience with this population is more limited, it appears that patients with DYT6 dystonia experience a generally good but more modest and less sustained response to pallidal DBS as compared to DYT1 patients [76‒78].

Myoclonus-dystonia (DYT11) is a syndrome characterized by both dystonia and myoclonic jerks, associated with mutations or deletions in the epsilon-sarcoglycan gene in about 30–50% of cases [79]. Most of the literature regarding DBS outcomes in patients with DYT11 myoclonus-dystonia is based on relatively small case series, but meta-analyses have demonstrated substantially positive outcomes in pediatric cohorts, with average and median motor improvements of dystonic components of around 70% [80, 81]. Improvement in myoclonus has also been documented [30, 82]. A combination of pallidal and thalamic stimulation has also been explored and has shown sustained benefit of motor symptoms in DYT11 cases [83].

Several case reports and small series have described DBS outcomes in pediatric patients affected by the newly identified DYT28 dystonia, caused by a mutation in the KMT2B gene. The spectrum of clinical manifestations includes generalized dystonia, in most cases involving the lower limbs at onset with later generalization, and unlike DYT1, spreads to include the larynx and oromandibular region as well [33]. In one series of adult and pediatric patients, 8 patients who underwent DBS had a median 38.5% improvement of BFMDRS motor score [33]. Another study with 15 adult and pediatric patients showed >30% motor improvement at 1 year in 8 patients, but this was sustained only in 5 patients. Dystonia improvement was maintained for trunk and neck more than oromandibular regions [84].

Despite the presence of an active neurodegenerative process and thus a reduced expectation for complete and sustained reversal of deficits, dystonias with central nervous system pathology have been targeted due to the lack of other treatment options. Most information about DBS outcomes is available for neurodegeneration with brain iron accumulation, which describes multiple degenerative diseases characterized by various movement disorders and cognitive decline, resulting in severe disability and shortened survival. The most common form is pantothenate kinase-associated neurodegeneration (PKAN), an autosomal recessive disorder caused by PANK2 gene mutations. While it can have variable phenotypes, the classic form causes childhood dystonia, ataxia, spasticity, parkinsonism, and cognitive impairment [85]. Due to its severity, PKAN has been a target of DBS therapy since the early years of therapy development [86], providing a large caseload with inconsistent outcomes. One meta-analysis noted short-term median improvement in the BFMDRS motor score of 27.7% in 36 pediatric patients with PKAN [80].

Lesch-Nyhan disease is an X-linked recessive disorder caused by mutations in the HPRT1 gene. Decreased hypoxanthine-guanine phosphoribosyl transferase activity causes excessive uric acid accumulation, which results in generalized dystonia, self-injurious behavior, intellectual disability, gout, and nephrolithiasis [87]. This rare and disabling condition has been treated with pallidal DBS in few selected cases, often targeting bilateral anteromedial in addition to posteroventral GPi in order to affect limbic symptoms with the modified stimulation target [34, 88]. A meta-analysis of 4 patients showed a mean of 26.4% improvement in the BFMDRS motor score [80]. Another series of 4 patients showed only a 3.4% mean improvement in BFMDRS motor score, but 2 patients showed >50% improvement in frequency and severity on behavioral scales, with 1 patient being able to return to school. These patients also had significant improvements in other disability and quality-of-life metrics [34].

Patients with acquired dystonia comprise a heterogeneous group, with multiple etiologies and neurologic deficits. These patients tend to have a more static pathological process than those of heredodegenerative pathology, but frequently present other issues – such as spasticity – that may limit outcome expectations after DBS surgery [5]. Cerebral palsy is the most common indication for DBS in secondary dystonia. Its prevalence is 2 per 1,000 live births [89], and underlying causes include hypoxic ischemic encephalopathy and prematurity. Overall, DBS short-term outcomes show variable benefits, ranging from 9.3% to 28.5% in mean BFMDRS motor score improvement [20, 22, 25, 28, 90, 91].

Many cases of focal or segmental dystonia fall into this category, mostly with adult onset, sporadic or familial occurrence. Patients with idiopathic dystonia have been treated with surgery in a similar fashion to those with DYT1 and other inherited forms of the disease. In pediatric patients without a recognized genetic mutation, those who have an isolated dystonia respond quite well to GPi DBS, with average improvements ranging from 31.7% to 72% as measured by the BFMDRS motor score and 26.4–66.5% as measured by the BFMDRS disability score [18‒20, 26, 73, 74].

In the following sections, we present data relative to long-term motor and neuropsychiatric outcomes of DBS, as well as complications including battery changes. In addition, we will review long-term outcomes after DBS placement for status dystonicus (SD).

We identified 20 articles that included individual data documenting motor outcomes of pallidal and subthalamic stimulation with 5 or more years of follow-up (Table 2). Data was available on a total of 78 individual patients of pediatric age; 43 were male, 34 were female, and 1 was unknown. The mean age at dystonia onset was 7.2 ± 3.3 years and mean age at DBS was 13.1 ± 3.9 years, with an average duration of disease prior to DBS of 6.1 ± 3.9 years. The average length of follow-up for the cases included in this study was 8.5 ± 3.3 years. The vast majority of patients (n = 69) were implanted within the GPi, with 8 patients receiving STN implantation and 1 patient targeted within the ventrolateral posterior nucleus of the thalamus. Dystonia was generalized in 77 patients, with only one focal dystonia [26]. This population included 58 patients with an inherited, isolated dystonia (42 DYT1, 7 DYT6, 7 DYT28, and 2 DYT11, shown in Table 2), 6 patients with heredodegenerative dystonia (4 PKAN and 2 Lesch-Nyhan), 2 patients with an acquired dystonia due to CP, and 12 patients with idiopathic dystonia. The average BFMDRS motor score at baseline was 57.8 ± 23.6 points. The baseline disability score, available in only 42 patients, was 17.8 ± 7.5 points. The mean percentage motor score improvement at last follow-up was 56.7 ± 40.9, with an average percentage disability score improvement of 61.4 ± 35.5%. These results support sustained motor benefit and reduced disability after DBS for dystonia.

Differential improvements among the different genetic subtypes of dystonia are shown in Figure 1. Patients with idiopathic, DYT11, and DYT1 dystonia demonstrated the greatest benefit, while patients with cerebral palsy and DYT6 showed moderate benefit, and patients with DYT28, PKAN, and Lesch-Nyhan showed the least long-term benefit. These results are consistent with previous studies with shorter follow-up showing greater benefit in patients with inherited, isolated dystonias and idiopathic dystonias compared to both heredodegenerative and acquired dystonias [80, 92]. Interestingly, the patients with acquired dystonia that emerged from our analysis had better outcomes than previously reported [20, 22, 25, 28, 90, 91], although they were still poorer overall as compared to patients with inherited, isolated dystonia. Poorer outcomes may be attributed to the higher incidence of structural brain damage in these patients, as well as disability due to other neurologic deficits and progressive neurodegeneration [80, 93].

Analysis of body regions involved by dystonia – available in 37 patients – showed 23 with facial and/or laryngeal involvement, 26 with cervical involvement, 32 with upper extremity involvement, 31 with lower extremity involvement, and 34 with truncal involvement. Out of 33 patients with data regarding region of onset, 18 had a lower extremity, 13 upper extremity, 1 laryngeal, and 1 cervical onset. Data regarding medication changes post-DBS was available for 16 patients, 13 of which (76%) were able to eliminate medications completely, while 3 patients were not. In 13 patients, dystonia spread to new regions post-DBS, including 7 spreads to the face, 4 to the larynx, and 6 to other regions. Given the prevalent spread of dystonia in pediatric patients [94], it is intriguing to notice that 22/35 (62%) of patients did not have spread to other regions after DBS.

There is a lack of data regarding long-term outcome predictors in pediatric dystonia patients that undergo DBS surgery. Overall, in primary generalized dystonia, age at onset [80, 93, 95], shorter disease duration [6‒8, 93, 96] and absence of fixed skeletal deformities are associated with better outcomes [6, 37]. Thus, DBS should be offered early in order to maximize benefit and avoid the development of musculoskeletal deformity, both in primary and secondary dystonia [8, 22]. Truncal and limb dystonia tend to respond better than orolingual symptoms [16, 80]. DBS is associated with a significant decrease in medication requirements and sometimes in complete elimination of dystonia medications [21]. Limited series have reported long-term benefits in pediatric cohorts, at 10 years and beyond [35, 73].

Literature regarding neuropsychiatric and quality of life outcomes in the long term is quite limited. In a STN DBS study including 9 pediatric patients with isolated dystonia and 10 years of follow-up, two reported stimulation-induced cognitive impairment with regard to executive function, one reported anxiety, two reported depression, and one reported weight gain. At the same time, all of them reported a significant improvement in quality of life. Gains included the ability to walk freely without assistance, perform ADLs independently including eating and writing, and function normally with return to school and work [35].

Another study included 2 pediatric patients with myoclonus-dystonia who underwent GPi DBS. Cognitive and quality of life scales were performed at the last follow-up, 5.3 and 5.7 years after implantation. They both had high scores on cognitive testing (MMSE), quality of life scales (SF-36), and anxiety/depression scales (BDI, HADS), with mild functional impairments on the social adjustment scale. One patient did report some mild anxiety, as well as some difficulty with social and physical role functioning, body pain, vitality, and mental health. Ultimately, these quality-of-life scores were similar to those of the general population. Furthermore, they were able to obtain their driver’s licenses as well as pursue higher level education [97].

In a cohort of 37 pediatric patients and adults, quality of life improvements persisted for the past 10 years, with low prevalence of cognitive impairment [98]. Another mixed cohort of 46 patients with a median follow-up of 4 years demonstrated complete elimination of dystonia medications in 61% of patients, and discontinuation of at least 1 class of medication in 91% at last follow-up [38]. A pediatric group with a median follow-up of 8 years demonstrated a decrease in mean daily anticholinergic dose from 29.9 (±35.6 mg) to 3 (±5.1 mg), also without any cognitive decline in MOCA scores at last follow-up [99].

SD is characterized by increased severity of dystonia requiring urgent hospital admission, often to an intensive care unit, due to life-threatening complications including bulbar weakness, respiratory failure, and metabolic derangements [100]. SD may occur in both primary and secondary dystonia, and is commonly triggered by infection or medication changes, or an interruption in chronic DBS treatment, although sometimes no precipitating factor may be found [101, 102] Refractory SD cases may require consideration of DBS and it has been argued that DBS should be considered early, particularly in conditions that are highly responsive such as DYT1 dystonia. However, more patients presenting with SD have secondary dystonia rather than primary dystonia [102]. DBS surgery was the most successful strategy in 89 adult and pediatric cases of SD [101]. In one cohort of 58 patients of all ages with SD who received DBS, about 90% of episodes resolved [103]. Interestingly, most patients in both series had secondary dystonia, suggesting that DBS should certainly be considered in all types of dystonia.

Five publications described a total of 9 pediatric patients with 5 or more years of follow-up after bilateral GPi DBS placement for the treatment of SD (shown in Table 3). The median age at time of DBS was 11 years (range 8.2–14.2) with a median of 30 days (range 14–180) after SD onset. The median time to SD resolution after DBS placement was 30 days (range 7–90) and the median follow-up was 96 months. Only 2 patients had isolated dystonia; the remaining seven had secondary or heredodegenerative dystonia. Motor scores were not available for all patients, but all of them had resolution of SD. Two patients maintained improvements in BFMDRS motor and disability scores with long-term follow-up. Two patients had initial motor improvements post-DBS but ultimately worsened compared to baseline at last follow-up. One patient gained the ability to walk independently 2 years post-DBS.

One particular risk in children when considering DBS is that they will likely require many battery changes over their lifetime, which may result in a higher risk of surgical complications compared to adults [104]. The mean battery life was 3.1 years in a cohort of children and adults with 9 years of follow-up [72]. The development of rechargeable IPGs has made this issue less important. Rechargeable devices have been shown to be well tolerated in children, with a low complication rate [105].

Adverse effects associated with DBS therapy are typically classified into three categories: surgical, hardware-, and stimulation-related. Typical short-term adverse events include post-op wounds or hardware infections. Risk of infection tends to be higher in the pediatric population for several reasons, including smaller body habitus, which results in delayed healing of the IPG incision, as well as strain due to severe dystonia which is more common in pediatric patients [38]. However, the frequency of these events decreases significantly over time. Therefore, long-term effects are typically hardware- or stimulation-related [106].

There is limited data regarding long-term adverse events in pediatric DBS patients. One study reported adverse events associated with DBS ranging from 1 to 15 years (65.8 months on average) after surgery in 42 patients, the majority of which were hardware-related. A total of 42 adverse events were described in 21 patients, 35 of which required surgical intervention. These included 6 accidental arrests of the IPG, 14 unscheduled IPG replacements due to technical defect or dislocation (after a mean of 73.7 months), 14 lead extension replacements due to high impedances, lead fracture or migration, shortening due to growth, or discomfort or pain (after a mean of 62.6 months), and 7 DBS system removals due to high impedances, dislocation, or malposition of the DBS electrode (after a mean of 56.6 months) [107]. While interruptions in stimulation have been associated with motor deterioration and even the development of SD [101, 102], there have been several reports of sustained relief despite prolonged interruption [108‒110]. Other limited reports of late adverse events include lead migration into the temporal lobe causing seizures at 7 years post-op, as well as electrode infection requiring removal and reimplantation at 9 years [26, 77]. Early studies suggested that lead fractures, particularly along the course of the extension cables, were more frequent in dystonia patients than in other movement disorder patients [12]. Later studies did not confirm this finding, although changes in surgical technique (i.e., the connector being placed above the mastoid rather than below) may have contributed to the reduction in lead fractures [37, 46, 111].

Manifestations of parkinsonism, including bradykinesia and freezing of gait, have been described after chronic pallidal stimulation for dystonia [112‒114] and may resolve upon using more dorsal contacts [113]. They increase in frequency with age, particularly after the fourth decade [98] and are therefore quite rare in the pediatric population. However, one study in DYT28 pediatric patients treated with GPi DBS reported freezing of gait at 3–6 years post-DBS [84]. In addition, stimulation-induced speech abnormalities, including dysarthria, dysphonia, and stuttering, are frequently reported [106]. Speech difficulties do complicate DBS therapy in the pediatric population, depending on stimulation intensity, likely related to activation of fibers of the internal capsule [107, 115].

Our systematic review of the literature describing long-term DBS outcomes in pediatric dystonia delivers an encouraging message, as motor and quality of life benefits of DBS for dystonia are maintained 5 or more years after implantation, with minimal cognitive or psychiatric deterioration. Serious complications are rare and the development of rechargeable IPGs has made frequent battery replacement less of an issue. Despite tremendous advances in genetic and molecular studies, which may ultimately result in a cure for dystonia, these results confirm the leading role of DBS in the current treatment options for most cases of generalized, medication-resistant, and disabling dystonia.

C.M. has no conflicts of interest to declare. M.T. received consultation fees from Abbot, Boston Scientific, and Medtronic. None of these relationships had any influence on the content of this review.

M.T. received funding from the Caron and Steven D. Broidy Chair in Movement Disorders.

C.M.: literature review and initial manuscript draft. M.T.: literature review and manuscript editing.