Effectiveness of Oral Sensory-Motor Stimulation in Premature Infants in the Neonatal Intensive Care Unit (NICU) Systematic Review

The aim of this study was to identify and to assess the best evidence currently available on the effectiveness of oral sensory-motor stimulation in preterm infants in the neonatal intensive care unit. We performed a systematic review following the Preferred Reporting Items for Systematic Reviews (PRISMA) statements. The search was conducted using the Pubmed, Web of Science (WOS), PEDro and Scopus databases. Clinical trials were reviewed and PEDro rating scale was used to assess the methodological quality of these studies. Results: 1267 studies were found and 11 were relevant and included in this review. Improvements were obtained in achieving independent feeding, maturation of the sucking pattern, transition to full feeding, motor function and length of hospital stay in most studies. Conclusions: there is evidence to support the benefits of the use of oral sensorimotor stimulation to achieve independent oral feeding in preterm infants, thereby reducing their stay in the Neonatal Intensive Care Unit.


Introduction
According to the World Health Organization, approximately 15 million babies are born prematurely every year, 8% to 10% of them in industrialized countries [1].
Pre-birth factors [2,3] combined with premature birth risk factors [4][5][6] increase the risk of death, which is estimated at one million babies dying in the first year of life [4][5][6][7][8]. Those who survive, approximately 50%, have developmental functional diversity such as motor, cognitive and behavioral impairments. More specifically, they may experience feeding difficulties, infections, jaundice, apnea of prematurity and retinopathy of prematurity [1]. Moreover, some of them are not limited only to the perinatal period, but they can be extended throughout life generating great disability and impact on the well-being of the babies [7][8][9][10].
In order to avoid these complications as much as possible and to ensure the extrauterine development during the last weeks of pregnancy, we have the Neonatal Intensive shorter time and a reducing length of stay in the hospital of preterm babies. In contrast to these articles, the present review aims to establish all the variables on which oral sensorymotor stimulation has an impact in premature infants admitted to the NICU.
The aim of this review is to identify and to evaluate, critically and objectively, the evidence currently available on the effectiveness of oral sensory-motor stimulation in preterm infants in the NICU.

Study Design
A systematic review was conducted following the Preferred Reporting Items for Systematic Reviews [PRISMA] [31] and Meta-analysis between January and March 2020. The databases PubMed, Scopus, Web of Science [WoS] and Physiotherapy Evidence Database [PEDro] were consulted.
This revision has been registered in PROSPERO with the code: 226833.

Database
Total Search Strategy 3. Advanced Search. "Premature" for title and abstract, "Stretching, mobilisation, manipulation, massage" for Therapy, "Head or neck" for Body part y "Clinical trial" for Method. PEDro n = 6 n = 0 n = 0 n = 0 Exclusion criteria were: (a) Population: Study subjects are in the family home with medical discharge. (b) Intervention: The intervention method of the study population included non-nutritive sucking but not oral/mucosal sensorimotor stimulation.
From each study, the variables related to the publication, as authors and year of publication, were obtained and reviewed. Variables related to the research carried out: sample size; age, sex and inclusion and exclusion criteria of the participants; main effects studied. In addition, we investigated the variables that were used to measure the effects and the applied intervention programs.
The variables evaluated in the different studies were those related to feed intake, volumes of feed consumed or lost, type of feeding, type of suction and growth control variables.

Assessment of the Methodological Quality of Studies
The methodological quality of the articles selected for this review was assessed using the PEDro scale [33]. This is a scale for the methodological assessment of randomised clinical trials involving physiotherapeutic intervention.
To guarantee the results, the assessment was performed independently by two investigators. Finally, the results were crosschecked. In case of discrepancy, a third investigator intervened.

Data Analysis
The meta-analysis of the data was performed using Review Manager 5.3 (The Nordic Cochrane Centre, The Cochrane Collaboration, (2014) [34] and EPIDAT software version 3.1 [35]. The first was used to check the homogeneity of the studies. Values were applied: I2 > 50%, indicating substantial heterogeneity, where randomized effect models were applied, and I2 < 50%, indicating substantial homogeneity, where the fixed effect model was applied. It was also used to obtain the results of the meta-analysis and the tables and graphs of forest and funnel plots. The second tool was used to check the risk of publication of the studies selected for meta-analysis. Whenever possible, Begg and Egger values were observed (p < 0.05).

Results
Out of 1267 studies, 11 studies were selected for analysis in this review. The number of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions, are reflected in the flow chart ( Figure 1).

Results
Out of 1267 studies, 11 studies were selected for analysis in this review. The number of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions, are reflected in the flow chart ( Figure 1).

Data Extraction
The sample size of the different studies ranged from 19 to 75 participants. In total, the cumulative sample under study was 507 subjects. The sex of the participants, not all studies specify it, but in those that do, we found that the number of boys under study was 242, compared to 162 girls. The age range was between 26 and 34 weeks of gestation. The age of the babies at the start of the intervention protocol corresponds to the gestational age at birth.

Data Extraction
The sample size of the different studies ranged from 19 to 75 participants. In total, the cumulative sample under study was 507 subjects. The sex of the participants, not all studies specify it, but in those that do, we found that the number of boys under study was 242, compared to 162 girls. The age range was between 26 and 34 weeks of gestation. The age of the babies at the start of the intervention protocol corresponds to the gestational age at birth.
Regarding the inclusion criteria, all infants were within the prenatal age range (less than 37 weeks' gestation) and admitted to the Neonatal Intensive Care Unit. With the exception of two studies, [36,37] the rest specified that feeding at the time of initiation should be complete tube feeding.
Some studies [38][39][40] considered the absence of respiratory support at least 48 h before starting the intervention together with normal blood oxygen levels as inclusion criteria while those led by Fucile et al. [38,[41][42][43] considered it necessary that the size of the subjects be appropriate to their gestational age. Hwang et al. [37] did not consider any of these criteria, with the exception of prenatal age which is common to all, but decided to include those infants with insufficient feeding (<4 mL of milk in the first 5 min). On the other hand, Ghomi et al. [36] established physiological stability at the time of oral stimulation, Apgar score of 6 and parental consent to participate as inclusion criteria for their study. Infants with congenital abnormalities or chronic disease were excluded from all studies. Information about the characteristics of the sample is given in more detail in Table 2.
To achieve independent feeding, different aspects were assessed: Overall intake (volume ingested/volume administered) and volume of milk taken during the first 5 min, volume of milk consumed in relation to session duration (mL/min) and volume of milk lost during feeding to assess feeding competence; [42] this ability was assessed by measuring the total volume of milk per spoon [39]. Other studies decided to establish two measures: the number of days it took for infants to achieve 30% independent oral feeding for the first 5 min, and the number of days it took to achieve 100% independent feeding, which they quantified as the oral milk intake (150 mL/kg/day) for three consecutive days [44]. Finally, the percentage of the prescribed volume ingested, mean volume ingested per suck (mL/suck) and time to complete feeding in minutes were measured [37].
Different parameters were chosen to assess the state of sucking ability in preterm infants: sucking frequency [37,42]; the measurement of non-nutritive sucking pressure through a dummy connected to a catheter [40] and the coordination between suckingsucking and breathing, as well as the level of maturity of sucking pattern and skill development, measured with the Lau scale [42,45]. The transition time from tube feeding to independent feeding was also extensively studied [36,42,46]. Some researchers specify that they used a wati spoon [38] and a spoon [39] as a measuring tool.
The measurement of infant growth in relation to oral stimulation through weight gain was considered by three studies. Measuring the value at the end of the study [36,38] or daily [43].
The impact of the intervention programme on the motor function of preterm infants was measured using the Test of Infant Motor Performance [TIMP] [38,43] and the Neonatal Oro Motor Assessment Scale [NOMAS] [43].
The effects on the number of days of hospital stay of infants in the NICU before discharge were also widely considered [36,38,43,44,46].
As for the oral sensorimotor intervention programs, these were variable in terms of the number of daily sessions, the time of application per session, the organization and method of application, and the overall application times (Tables 2 and 3).

Assessment of the Methodological Quality of Studies
The total score on the PEDro scale ranged from 4 to 8 points. According to the results obtained, it can be seen that the greatest methodological deficit corresponds to the items referring to blinding of the therapists, hidden allocation of subjects and analysis by "intention to treat", as in, almost none of the studies evaluated was a positive response obtained. Most of the clinical trials analysed scored 5/10 or less (Table 4).

Data Analysis and Outcomes
Of all the variables extracted from the studies included in this research, only two were included in the meta-analysis.
The results of this meta-analysis indicate that there is only heterogeneity in the duration of hospital stay. In addition, statistically significant results are observed in favour of the intervention group in both variables (Figures 2 and 3).       The results of the Begg and Egger tests indicate that there is only a risk of publication in the variable number of days to reach full oral feeding (Table 5). These results are confirmed by the funnel plots (Charts 1 and 2).

Variable Begg (p) Egger (p)
Duration of hospital stay p = 0.0085 p = 0.0003 Number of days to reach full oral feeding p = 0.7341 p = 0.2092 Figure 3. Forest plot panel of the comparison of the difference between the groups in number of days to reach full oral feeding.

Discussion
The systematic review of studies and meta-analysis of variables indicate that oral sensorimotor stimulation in the management of preterm infants hospitalized in the Neonatal Intensive Care Unit [NICU] is beneficial.
All of the reviewed studies have obtained results that reflect the benefits of oral sensorimotor stimulation in the management of premature infants. Although it is true that not all of them have studied its effects on the same variables, it is also true that not all of them have studied its effects on the same variables.
The meta-analysis shows that the intervention oral sensory-motor stimulation is effective in premature infants in the NICU. Although only two variables could be compared, they show benefits of this type of intervention in terms of reduction of hospital stay and Chart 2. Funnel Plot of number of days to reach full oral feeding.

Discussion
The systematic review of studies and meta-analysis of variables indicate that oral sensorimotor stimulation in the management of preterm infants hospitalized in the Neonatal Intensive Care Unit [NICU] is beneficial.
All of the reviewed studies have obtained results that reflect the benefits of oral sensorimotor stimulation in the management of premature infants. Although it is true that not all of them have studied its effects on the same variables, it is also true that not all of them have studied its effects on the same variables.
The meta-analysis shows that the intervention oral sensory-motor stimulation is effective in premature infants in the NICU. Although only two variables could be compared, they show benefits of this type of intervention in terms of reduction of hospital stay and in number of days to reach full oral feeding. However, the implementation methodology is very heterogeneous. This prevents the best method of stimulation intervention is known. Even so, those studies that used the PIOMI, refs. [37,38] as well as being the most current, demonstrate a significant improvement in motor function compared to the rest of the studies. This leads us to believe that PIOMI may be the best intervention to improve motor function in preterm infants.
Our results are consistent with previous studies [14,20,22,25,47]. We also observed that these benefits also occur with other types of therapy [20,28,48]. This implies that treatment through stimulation and touch is essential for babies admitted to the Neonatal Intensive Care Unit, as long as they do not expose babies to stress [10].
All studies that evaluated feeding parameters reported early acquisition of independent breastfeeding and improvements in sucking patterns, with the exception of suckingswallowing coordination, where the results obtained were not considered significant [46]. These results should be complemented by other findings such as maturation in the developmental stages of sucking, as these are fundamental to understanding possibilities for improvement in the infant's sucking skills [49]. In addition, it would be interesting to know how the combination of oral sensorimotor stimulation is complemented by other types of massage that improve gastric motility [13,48]. Of the studies that evaluated hospital stay, three [37,38,44] showed a significant decrease, while the other two [43,46] could not be considered relevant.
Variables classified as infant alertness, physiological data and breastfeeding skills did not achieve significant changes in the studies where they were evaluated [37,46].
As for the differences in the time and frequency of the sessions, there are no data that demonstrate greater benefits of one or the other. Even so, all the studies report positive results in the evaluation of the different variables. This indicates that this therapy will always be a reason for improvement in the development of premature infants.
The assessment of the long-term effects of the intervention plan was measured in only one study, [46] confirming that the subjects continue breastfeeding satisfactorily. It would be interesting for future studies to consider assessing this aspect. This would support that the benefits of the therapy do not only work in the short term, but are sustained over time.
Of the studies analysed and included in this review, none showed any type of contraindication or negative side effects during the intervention programme or afterwards once it has been completed. Furthermore, it is worth highlighting the low cost of this type of intervention programme, due to the fact that specific devices would not be needed for its implementation, but rather the figure of the physiotherapist. This would result in an improvement in the quality of care for this type of patient, without a high economic impact on the Neonatal Intensive Care Units, if the professional is already on staff.
Research to date on oral sensorimotor intervention as a strategy in the management of preterm infants is quite limited. In addition, there is a need for a greater variety of authors to conduct this type of study. The characteristics of the stimulation program were not specified in all studies. This makes it difficult to know which programmes have been applied and prevents replication of those with the best results.
No reviews have been found that specifically refer to the impact of oral sensorimotor stimulation on these babies. Our study appears to be the first to bring together the current literature on this type of approach. Therefore, it can be a tool for guidance and consultation on the different protocols of oral sensorimotor therapy in the treatment of preterm patients admitted to the NICU.

Conclusions
Oral sensorimotor stimulation is beneficial for the acquisition of independent oral feeding in preterm infants by decreasing the days of admission and the number of days to reach full oral feeding.