Effect of oromotor exercises on feeding in children with cerebral palsy: systematic review

Feeding problems are prevalent in children with cerebral palsy (CP). Oromotor exercises (OME) should be started as soon as possible to enhance chewing and drooling. Oromotor exercises consist of active exercises, passive exercises, and sensory stimulation. The purpose of this review is to evaluate the effectiveness of oromotor exercises on feeding, chewing, and drooling in children with CP. The American Academy for Cerebral Palsy and Developmental Medicine and Preferred Reporting Items for Systematic Reviews and Meta-Analyses methodology were used to conduct a systematic review. Four databases (PubMed, Cochrane Library, PEDro, and Google Scholar) were searched; this review includes seven articles, participants were 173 participants ranging in age from 18 months to 18 years. Articles were assessed according to their level of evidence and quality assessment was done by AACPDM, PEDro scale, and JBI scale. Due to the heterogeneity across included studies, descriptive analysis was performed on all of them. Primary outcomes were chewing and drooling. Results showed the effectiveness of OME in improving drooling, but with weak evidence while not effective in improving chewing. High-quality studies are required to develop a firm judgment on the influence of oromotor exercises on feeding. The current level of evidence to support the effectiveness of oromotor exercises in children with CP is currently insufficient.

dysfunction that persists causes feeding difficulties, which leads to growth and development retardation [5]. Drooling, on the other hand, has a negative impact on social development and creates physical problems [6]. Children with CP have changes in their oral functions, which make feeding difficult and can lead to major health problems like malnutrition and pneumonia [7].
One of the elements affecting children's health is feeding efficiency. Feeding problems are prevalent in children with cerebral palsy, resulting in insufficient caloric intake and eventually malnutrition [8]. In children with moderate to severe CP, feeding difficulties are common accompanied with poor feeding status and poor health [9].
Pediatric feeding disorder (PFD) is defined as a lack of age-appropriate oral intake that is linked to medical, nutritional, feeding skill, and/or psychosocial problems [10]. Up to 25% of all children have feeding problems or dysphagia; prematurely born infants had a 40% chance of having swallowing problems, a 64-78% chance of having developmental problems, and a 99% chance of having CP [11].
Feeding problems and poor nutritional status are frequent in children with CP, especially as their gross motor impairment and age increase, and they may have an adverse effect on their health, physical development, and cognitive development [12]. Choking with food (56%), feeding duration more than 3 h per day (28%), frequent biting (22%), and chewing difficulty (26%) are all common feeding problems in children with CP [13].
Chewing is described as a rhythmic oral motor activity for communicating and softening solid food as part of the feeding process [14]. Food transportation from the front of the mouth to the molar area, where it is processed through several masticatory cycles, is the most affected aspect of chewing in patients with CP due to insufficient lateral and rotational tongue motions [15]. Drooling is an uncontrolled expulsion of saliva from the mouth that is frequent in children with CP [16]. Furthermore, these disorders can lead to a variety of health problems, including poor nutritional status and growth, reactive airway disease, and aspiration pneumonia [17].
Active exercises, passive exercises, and sensory applications are the three primary types of OME employed in clinical practice [18]. Active range of motion, stretching, and strength training are examples of active exercises; through the recruitment of new motor units as muscle fibers are expanded, strength, endurance, and power are all developed through these exercises [19].. Passive exercises include passive range of motion exercises, in which the movement is performed partially or wholly by the therapist or caregiver, with little or no participation from the person getting therapy. Sensory input is provided, circulation is improved, and joint flexibility is maintained or improved using these techniques [20]. The application of cold, heat, electrical stimulation, high-frequency vibration, or other agents to muscle tissues is referred to as sensory applications. Kumar et al. [21] reported a significant improvement in drooling and chewing after OME in children.
Despite the fact that there are numerous management alternatives, only a handful have been explored, including OME for drooling, there is not enough research to know how OMEs affect swallowing [22]. Arvedson et al. [22] in their systematic review recommended further research due to the dearth of studies with enough data keeping in view the stressful and socially isolating effects of drooling [22]. Three studies were taken from the systematic review of Arvedson et al. and new four studies were added to them. The purpose of this review is to examine the present research's quality of OME effectiveness on chewing and drooling in children with CP. There is an urgent demand for high-quality studies.

Methods
To find relevant published studies, we used PubMed, Cochrane Library, PEDro, and Google Scholar. The following keywords were used to search those databases: CP, Oral motor exercises, OME, Systematic Review, Oral sensorimotor, Chewing, and Drooling. This systematic review was carried out according to the criteria included in Cochrane Handbook for Systematic Reviews of Interventions [23], American Academy for Cerebral Palsy and Developmental Medicine (AACPDM) methodology for developing systematic reviews of treatment interventions [24], and Preferred Reporting Items for Systematic Reviews (PRISMA) guidelines [25].

Eligibility criteria
Studies that matched the criteria listed below were considered eligible:

Study design
Published all study designs studying the effect of OME in children with cerebral palsy except bench research, common sense/anecdotes, expert opinion, case study or report, non-randomized controlled AB single-subject research design (SSRD), and review articles. In this systematic review, two studies were randomized control trials [26,27], three studies were quasi-experimental [16,28,29] and two studies were case series [30,31].

Participants
This systematic review concerned children with the following inclusion criteria: CP, aged between 18 months and 18 years. No study determined the type of CP except two case series studies [30,31]. One study: subject 1 was spastic diplegia, subject 2 was spastic quadriplegia [30]. Other study: subject 1 was spastic CP, subject 2 was athetoid CP [31].

Type of intervention
Oromotor exercises in the form of active exercises, passive exercises, and sensory stimulation, duration between 10 days and 9 months.
Outcomes: Chewing (performance) and drooling (severity, frequency, weight of saliva, and percentage of drooling).
Language: Full-text articles are available in English.

Exclusion criteria
Studies were excluded if they were: review studies, bench research, common sense/anecdotes, expert opinion, case study or report and non-randomized controlled AB SSRD, studies on children other than CP, studies that measured outcomes unrelated to the study's objectives, unpublished studies, studies published in a language other than English, and studies that combined OME with other types of modalities.

Search methods for identification of studies
References were searched from 1980 up to December 2020, using the following electronic databases: PubMed, Google Scholar, the Cochrane Library, and PEDro. Search terms used the following keywords "oral motor exercises," "cerebral palsy," "OME," "oral sensorimotor," "chewing," and "drooling" and using Boolean operators AND/OR. To find the relevant research, the titles and abstracts were checked against the inclusion and exclusion criteria. When the abstract revealed eligible study, the full text was obtained for complete assessment.

Treatment procedures
The articles discussed OME's impact on chewing and drooling in children with CP; participants received OME as a control group in three studies [26][27][28], and as a study group of four studies [16,[29][30][31]. Exercises included traditional oral motor exercises which include active and passive exercises of the tongue and lips [26,27]; blowing candle; blowing balloons; blowing bubbles; sucking games [28]; techniques for sensory stimulation: face massage, tapping, stroking, brushing, and ice stimulation [29]; oral facilitation techniques [30]; oral motor stimulation then vibration [31]; and muscle vibration [16].

Data extraction and analysis
Data was extracted by two authors RF and AF and the third FH was consultant, according to data extraction form developed by AACPDM Treatment Outcomes Committee version 2008 [24]. Data was extracted according to the following items from the included articles: (1) the author and year of publication; (2) population information, including the numbers of children participating, their diagnosis, and their ages; (3) study design; (4) methodology, including the type of intervention, technique of its application, and its duration; (5) measured outcomes; and (6) results.

Level of evidence
The level of evidence of group design studies was scored according to Sacket et al. [32] and the level of evidence for the single-subject design was scored according to Logan et al. [33].

Results
After searching, 2437 articles were found (Fig. 1). PubMed recorded 326, PEDro recorded 7, Cochrane Library recorded 29, and Google Scholar recorded 2080. After removing duplicates, the records were 1284 studies. The records screened were 1284 while 1265 were excluded from abstract, then 19 full-text articles were assessed for their eligibility. Tweleve articles were excluded for various reasons because of different interventions, the children's diagnosis was not CP, or not the same outcome; finally, seven articles were included in this systematic review.

Characteristics of studies
Data was compared and findings were represented after extracting data from each study included in this systematic review. The variation among studies with regard to outcome measures, interventions, and methodological quality of the studies did not allow us to perform a quantitative analysis (meta-analysis). The existing data was not homogenous, so the current studies were analyzed by using descriptive analysis. Included studies are presented in Table 1.

Intervention
In this systematic review, oromotor exercises were defined broadly and a variety of interventions were examined as discussed before in the treatment procedure.

Comparators
Two studies applied Functional Chewing Training (FuCT) in the form of positioning the child, positioning the food, sensory stimulation, chewing exercises, and adjustment of food consistency [26,27] and one study applied OME in addition to chin cup, and another group with no treatment applied [28]. There were no other comparators because other studies were of a singlesubject research design [16,[29][30][31].

Types of outcomes measured
Although chewing and drooling were the outcomes in the present systematic review, other outcomes were measured in studies that were included, for example, feeding behaviors [26], tongue thrust severity [27], and speech [30]. Chewing was the outcome of two articles (randomized controlled trials) [26,27] and drooling was the outcome of six articles [27][28][29][30][31] and [16].

Measurement of chewing and drooling
In the present systematic review, chewing was measured by the Karaduman Chewing Performance Scale [26,27]. Drooling was measured by drooling severity and frequency scale, drooling impact scale, visual analogue scale, and drooling quotient [16], an OHAUS 700 series triple balance [30], momentary time sampling procedure [31], scales which weigh to the nearest 1 g [28], and drooling severity and frequency scale [27,29].
Level of methodological quality Table 2 shows the results of each study on the AACP DM. Both studies scored 6 [26,27]. The PEDro scale score for each study is shown in Table 3; one study had a score of 8 [26] and the other study was given a score of 5 [27]. Tables 4, 5, and 6 provide the results of all studies on the JBI scale. The score of randomized studies is presented in Table 4; one study obtained a score of 11 [26] and the other study had a score of 8 [27]. The score of quasi-experimental studies is presented in Table 5;  one study had a score of 7 [29], one study received a perfect score of 6 [16]; and one study obtained a score of 4 [28]. The score of case series studies is presented in Table 6; one study had a score of 7 [30] and the other study received a perfect score of 6 [31].
On the Cochrane risk of bias tool, the risk of bias for all studies is presented as follows: for random sequence generation, only two studies go into enough depth about the method utilized to create the allocation sequence to determine if it should produce comparable groups [26,27]; the other five studies are unclear in description [16,[28][29][30][31]. For allocation concealment, all studies are unclear in description [16,[26][27][28][29][30][31].
The last item which is other sources of bias states any important concerns about bias not addressed in the domains mentioned in the assessment of bias in studies; four studies fulfill this item [16,26,27,29] and only three studies were unclear in description [28,30,31].

Level of evidence
The level of evidence in each of the seven articles was evaluated in accordance with the AACPDM Treatment Outcomes Committee version 2008.
Two studies of group design were on level II [26,27], one study of group design was on level IV [28], and the other four studies of the single-subject design were on level IV [16,[29][30][31] as presented in Table 1. Oromotor exercises are effective in improving drooling, but with weak evidence and not effective in improving chewing.
The existing data was not homogenous, so the current studies were analyzed using descriptive analysis. Table 1 gives a general description of all the studies that were included, including the basic study design, level of evidence, characteristics of the participants, intervention, outcome measures used, and results.

Discussion
Searching in the literature revealed low evidence of the effectiveness of OME on feeding: chewing and drooling.  Oral-motor exercises are non-speech activities in dealing with CP children. One study reported that no adverse events were detected following this intervention, while the other six studies did not report it.
The small number of studies that met the inclusion criteria was the review's main limitation, meta-analysis not allowed because of different outcomes of interest, and the methodology of these research, as well as the type and duration of OME, all are different, so descriptive analysis was utilized.
In this systematic review 7 articles discussed the effect of OME on chewing and drooling, only two studies were RCTs [26,27], three studies were quasi-experimental [16,28,29] and two studies were case series [30,31].
Awan et al. [37] reported a more significant reduction in drooling severity and it's impact in the group who received kinesio taping and OME as compared to the other group who received kinesio taping p ≤ .05. Kumar et al. [21] reported a significant improvement in drooling and chewing after OME in children; this agrees with all studies which reported that OME were effective in the improvement of drooling in children with CP.
All outcomes of studies represent the ICF component of body structure and body function only except one study by Serel Arslan et al. [26] which included outcomes representing the ICF component of activity and participation and body structure and body function, the remaining ICF component of contextual/environmental factors were not mentioned in any study.
The study by Serel Arslan et al. [26] who applied traditional oral motor exercises for 12 weeks, including active and passive exercises of tongue and lips showed no  change was found in chewing. The study by Inal et al. [27] who applied classical oral motor exercises for 12 weeks, including active and passive exercises of the lips and tongue showed no improvement in chewing performance, drooling severity and frequency. The study by Harris and Dignam [28] who applied OME for 9 months, including sucking games, games of blowing out candles, blowing up balloons, and blowing trumpets showed improvement was 28%. The study by Fatima et al. [29] who applied oral motor exercises over a period of 6 months, including techniques for sensory stimulation; face massage, tapping, stroking, brushing, and ice stimulation showed a significant reduction in drooling p < 0.05. The study by Russo et al. [16] who applied vibration for 10 days showed statistically significant differences between baseline and (10 days, 1 month, and 3 months) p < 0.001 in all scales and no statistical significance between 10 days to 1 month, 10 days to 3 months, and 1 month to 3 months in all scales. The study by Iammatteo et al. [30] who applied oral facilitation techniques for 12 days, including gentle stroking and firm pressure around lips and inside the mouth in the context of play showed decreasing drooling and study by Domaracki and Sisson [31] who applied hourly treatment of oral motor stimulation, then vibration for 10 s, oral motor stimulation, including an NUK device which was used to stimulate the child's hard palate, brush the upper and lower gums, massage the center and both sides of the tongue, and make strokes to the inside of each cheek, showed decreasing drooling by oral motor stimulation, but when vibration was applied, it did not provide any additional therapeutic effects. This variation did not allow us to make a consistent conclusion about the best type and duration of OME to improve feeding. This systematic review showed the effectiveness of OME in improving drooling, but with weak evidence while not effective in improving chewing.
Given the small number of studies and high heterogeneity among them, caution is advised when interpreting the current findings. Future research is needed to learn how to maintain the positive effect of OME throughout time and to identify the important characteristics of OME (intensity, frequency, and duration).

Conclusion
This systematic review is an attempt to close the gap in knowledge that exists between research and clinical practice in using OME in children with CP. There is a clear need for more RCTs focusing on this issue to  establish strong evidence. The current level of evidence to support the effectiveness of OME in children with CP is currently insufficient. It could be concluded from the existing evidence that OMEs were effective in improving drooling. Finally, we can conclude that there is a significant gap between OMEs and research evidence.