Abstract
Assessment of both motor development and sensory processing can play an essential role in identifying infants who need early intervention, which occupational therapy practitioners can play a crucial role in.
The World Health Organization has defined preterm birth as birth before 37 completed gestational weeks, subdivided according to gestation: extremely preterm (<28 wk), very preterm (28–31 wk), and moderate–late preterm (32–36 wk; Howson et al., 2012). As gestation increases, the survival rate also increases and the morbidity rate decreases (Stoll et al., 2010). Preterm infants have a higher risk of developmental disorders, including motor, sensory processing, and cognitive disorders, than infants born at term (Bhutta et al., 2002 ; Bos et al., 2013; Cabral et al., 2016).
Technological advances in neonatal intensive care units (NICUs) have led to an increase in survival rates for preterm infants, and these advances have been accompanied by increased rates of neonatal morbidities and developmental disorders (Doyle et al., 2010; Stoll et al., 2010; Younge et al., 2017). NICUs are a stressful environment for preterm infants, who have a rapidly developing but immature brain (Bröring et al., 2017). Various sensory stimulations in the NICU are associated with negative physiological responses such as increased apnea, hypoxemia, increased intracranial pressure, and instability in arterial oxygen tension (Brown, 2009; Vitale et al., 2021). Smith et al. (2011) found that more exposure to stressors such as medical procedures and other stimuli in preterm infants in the NICU can cause a decrease in the size of the frontal and parietal lobes, using MRI, and alterations in motor behavior, as shown by neurobehavioral examination.
The sensory systems develop sequentially during intrauterine life (Lickliter, 2000). However, exposure of preterm infants in the NICU to sensory stimuli such as high frequency and sharp sounds, bright lighting, and unpredictable and painful touch during medical procedures alters this process (Graven, 2000; White-Traut et al., 1994). Touch is the first sensory system to develop, but it receives less stimulation in the NICU as a result of the isolation of preterm infants in incubators, and although the visual and auditory senses are the last to develop, these receive more stimulation in the NICU (White-Traut et al., 1994). Inappropriate sensory stimulation may cause sensory processing difficulties, resulting in developmental delays (White-Traut et al., 1994).
Chorna et al. (2014) found that abnormal sensory reactivity in preterm infants during the first year of life was associated with worse developmental outcomes at age 2 yr. Eeles, Anderson, et al. (2013) showed an association between sensory processing and development parameters in preterm infants at age 2 yr. Additionally, Ryckman et al. (2017) demonstrated that preterm infants were at higher risk of sensory processing difficulties at ages 4 to 6 yr than term infants. Kaya Kara et al. (2020) found that sensory processing was associated with motor development at corrected (postterm) age 1 mo and 4 mo in preterm infants, as indicated by Infant Sensory Profile–2 scores (Dunn, 2014). The Infant Sensory Profile–2 provides professionals with a way of documenting children’s sensory processing patterns from early life, and Dunn (2014) stated that scores on this instrument can be used to identify whether follow-up is necessary.
The first motor patterns, such as startle, occur at 7 wk gestation, followed by the general movements (de Vries et al., 1985; Piontelli, 2010). General movements, defined as movements of the entire body occurring in variable sequences of arm, leg, neck, and trunk movements (Einspieler & Prechtl, 2005), appear in term infants at 9 wk postmenstrual age and last up to 20 wk postterm age (Einspieler & Prechtl, 2005; Einspieler et al., 2004). General movements, which have been likened to storms in terms of both motion and sensorimotor functioning (Piontelli, 2010), cause sensory stimulation even in intrauterine life, before the brain receives sensory stimulation from outside stimuli (Fagard et al., 2018). Fagard et al. (2018) pointed out that all spontaneous movements, including the general movements, play an essential role in the development of bones, cartilages, ligaments, tendons, and muscles (Müller, 2003) and are important in the development of sensorimotor systems (Fagard et al., 2018).
General movements occur in age-specific patterns, and at postterm ages 3 to 5 mo they are called fidgety movements (Einspieler et al., 2004; Prechtl et al., 1997). With regard to predictive value, Burger and Louw (2009) reported that measures of fidgety movements can be used to identify infants with neurodevelopmental disabilities with a sensitivity of ≥92% and a specificity of ≥82%. In addition, the General Movements Assessment (GMA; Einspieler et al., 2004) provides global and detailed scores that can be used to assess infants’ concurrent repertoire of movements and postures. The Motor Optimality Score (MOS) of the GMA was found to be associated with preterm infants’ postural and motor control (Peyton et al., 2016), with cognitive development in children born preterm (Butcher et al., 2009), and with self-mobility in children later diagnosed with cerebral palsy (Einspieler et al., 2019; Yang et al., 2012).
To the best of our knowledge, no study has investigated the development of early spontaneous movements and sensory processing in preterm infants before corrected age 20 wk. We sought to answer two questions: (1) Do global and detailed GMA results for early spontaneous movements and Infant Sensory Profile–2 results for sensory processing differ by gestational age at corrected age 3 mo to 5 mo? (2) Are MOS and Infant Sensory Profile–2 results related to each other in the first months of life in preterm infants?
Method
Participants
This prospective cross-sectional study included 111 preterm infants with a corrected age of 9 to 20 wk; 50 were born at <32 wk gestation (24 female, M age [SD] = 13 wk, 3 days [1 wk, 6 days]) and 61 at 32 to 36 wk gestation (32 female, M age = 13 wk [1 wk, 6 days]). These infants had been admitted to the Faculty of Physical Therapy and Rehabilitation, Developmental and Early Physiotherapy Unit of Hacettepe University in Ankara, Turkey. We followed these infants and routinely assessed (1) their early spontaneous movements using the GMA and (2) their sensory processing using the Sensory Profile–2. Inclusion criteria were as follows: gestation of <37 wk, stay in the NICU of ≥1 day, and corrected ages 9 to 20 wk; those with any congenital anomaly were excluded.
We performed a post hoc power analysis using G∗Power Version 3.1 for the effect size and power of analysis. Effect size was .739 using the MOS results of all infants. Power was 96% with an error rate of α = .05.
Procedure
We recorded videos of the infants’ early spontaneous movements at a median corrected age of 13 wk (range = 9–19 wk). Three- to 5-min video recordings were made with the partly dressed infant in a supine position during active wakefulness, consistent with GMA instructions (Einspieler et al., 2004). Video recordings were made and the Infant Sensory Profile–2 was administered on the same day in a single session. All research procedures were approved by the Non- Interventional Clinical Research Ethics Board of Hacettepe University. We obtained written informed consent from the parents of all infants before assessment.
Measures
General Movements Assessment
Two certified scorers (advanced level; Akmer Mutlu and Ayşe Livanelioğlu) who were not familiar with the infants’ clinical histories separately assessed fidgety movements as well as concurrent movements and postures from infants’ video recordings using the GMA. In cases of disagreement (4 recordings; 3.6%), the video recordings were jointly discussed between the two scorers until consensus on scoring was reached.
For the MOS, the scorers used the revised score sheet for infants ages 3 to 5 mo (Prechtl’s method; Einspieler et al., 2019); the MOS ranges from 5 = worst performance to 28 = best performance (Einspieler et al., 2004). The MOS consists of scores (in parentheses) in five subcategories: Temporal organization and quality of fidgety movements: normal (12), abnormal exaggerated (4), or absent (1) Observed movement patterns other than fidgety movements: predominantly normal (4), equal number of normal and abnormal patterns (2), or predominantly abnormal (1) Age-adequate movement repertoire: adequate (4), reduced (2), or absent (1) Observed postural patterns: predominantly normal (4), equal number of normal and abnormal patterns (2), or predominantly abnormal (1) Movement character: smooth and fluent (4); monotonous and/or jerky, stiff, tremulous, slow/fast (2); or cramped–synchronized (1) (Einspieler et al., 2004, 2019).
Interscorer reliability (intraclass correlation coefficient) for the MOS ranges between .80 and .94 (Fjørtoft et al., 2009).
Infant Sensory Profile–2
The Sensory Profile–2 is a revised version of the Sensory Profile assessment family, which includes the Sensory Profile, the Infant/Toddler Sensory Profile, the Sensory Profile School Companion, and the Sensory Profile Supplement (Dunn, 2014). The Sensory Profile–2 is a parent-report questionnaire, and the Infant Sensory Profile–2 is the part of the Infant/Toddler Sensory Profile–2 that documents infants’ sensory processing patterns from birth to age 6 mo. The Infant Sensory Profile–2 contains 25 items yielding a total raw score. Dunn (2014) developed a normal curve– based classification system for the Sensory Profile–2: Raw score totals are classified as much less than others (lowest 2%), less than others (1–2 SD below the mean), just like the majority of others (<1 SD above or below the mean), more than others (1–2 SD above the mean), and much more than others (highest 2%). We classified participants who scored just like the majority of others as having typical sensory processing and the rest as having atypical sensory processing. The reliability and validity of the Infant Sensory Profile–2 have been reported in detail elsewhere (Dunn, 2014; Kayıhan et al., 2018). The first author (Bilge Nur Yardımcı-Lokmanoğlu) administered the Infant Sensory Profile–2 in face-to-face interviews with parents.
Statistical Analysis
We used IBM SPSS Statistics (Version 25) for statistical analyses. Nonparametric tests were used when the variables were not normally distributed. A p value of <.05 was considered significant for all tests. We compared demographic and clinical data between groups using Pearson χ2 for categorical variables (e.g., sex, MOS subcategory results) and Mann–Whitney U for continuous variables (e.g., gestational age, MOS results, Infant Sensory Profile–2 raw scores). For analysis of relationships, we used the Spearman correlation coefficient, classified as follows: coefficients of .90 to 1.00 indicate a very high correlation, .70 to .89 a high correlation, .50 to .69 a moderate correlation, .30 to .49 a low correlation, and .00 to .29 little or no correlation (Asuero et al., 2006).
Results
The infants’ gestation at birth ranged from 23 wk to 36 wk (median = 32 wk), and their birth weight ranged from 570 to 4,100 g (median = 1,770 g). Table 1 presents the demographic and clinical characteristics of the infants and parents.
Infant and Parent Demographic and Clinical Characteristics
Pearson χ2.
Mann–Whitney U.
Total serum bilirubin value >12.9 mg/dl.
Infants with Grade III and above were included.
According to SARNAT classification, infants with Grade II and above were included.
Between mothers.
Between fathers.
Missing data for 3 mothers and fathers in each group for educational level and employment status.
p < .05.
Early Spontaneous Movements
Table 2 presents results for the infants’ early spontaneous movements.
Motor Optimality Scores at Corrected Age 3–5 Mo
Mann–Whitney U.
Pearson χ2.
p < .05.
Infants Born at <32 Wk Gestation
Of the 50 infants born at <32 wk gestation, 72% had normal fidgety movements and 28% had aberrant fidgety movements (13 absent, 1 abnormal [exaggerated]). The quality of other movement patterns was predominantly normal in 82% of the infants and predominantly abnormal in 14%, and 2 infants (4%) showed an equal number of normal and abnormal patterns. The most frequently occurring abnormal movement patterns were mouth movements (18%), decoupled kicking (14%), and tongue movements (10%).
Fifty-four percent of the infants born at <32 wk gestation were able to maintain the head centered, 40% had symmetrical body posture, and 80% had variable finger postures. Twenty percent had a lack of variable finger postures, 14% showed predominant fisting, and 1 infant (2%) showed synchronized opening and closing of fingers.
Sixty-two percent of the infants displayed a monotonous and/or jerky, stiff, tremulous, slow/fast movement character. In addition, 4% had a cramped– synchronized movement character, and 34% showed a smooth and fluent movement character.
Infants Born at 32 to 36 Wk Gestation and Comparison Between Groups
Of the 61 infants born at 32 to 36 wk gestation, 98% had normal fidgety movements; only 1 infant (2%) showed absent fidgety movements. The presence of fidgety movements differed significantly between the groups (p < .001).
All infants born at 32 to 36 wk gestation showed predominantly normal movement patterns, a significant difference compared with the infants born at <32 wk gestation (p = .003). Furthermore, none of the abnormal movements were classified as >10%. The two groups did not differ significantly in age-adequate movement repertoire (p = .309).
The majority (71%) of the infants born at 32 to 36 wk gestation were able to keep the head centered, 54% had symmetrical body posture, and 87% showed variable finger postures. Only four infants (7%) had predominant fisting. No significant difference in posture was found between the groups (p = .114).
Like those born at <32 wk gestation, the majority of infants born at 32 to 36 wk gestation (67%) showed a monotonous and/or jerky, stiff, tremulous, slow/fast movement character. No significant difference in movement character was found between the groups (p = .277).
Sensory Processing
Table 3 presents data on the sensory processing of the infants. On the Infant Sensory Profile–2, 44% of the infants born at <32 wk gestation had atypical overall scores, whereas 20% of the infants born at 32 to 36 wk gestation had atypical overall scores (p = .006). However, no significant difference in Infant Sensory Profile–2 raw scores was found between the groups (p = .306).
Infant Sensory Profile–2 Results
Mann–Whitney U.
Pearson χ2.
p < .05.
Relationship Between Early Spontaneous Movements and Sensory Processing
Table 4 presents results for the relationship between early spontaneous movements and sensory processing. We found no relationship between results for the MOS and its subcategories and the Sensory Profile–2 raw scores and typical–atypical categories in the infants at 9 to 20 wk corrected age (p > .05).
Relationship Between Early Spontaneous Movements and Sensory Processing at Corrected Age 9–20 Wk
Note. MOS = Motor Optimality Score.
Spearman correlation coefficient.
p < .05.
Discussion
In this study, preterm infants born at <32 wk gestation demonstrated more abnormal early spontaneous movements and sensory processing difficulties than those born at 32 to 36 wk gestation. To the best of our knowledge, our study is the first to use the global and detailed GMA and the Infant Sensory Profile–2 to investigate the relationship between early spontaneous movements and sensory processing in the first months of life. Previous studies that evaluated motor development and sensory processing in infants aged <1 yr have conflicted regarding the relationship between motor development and sensory processing (Cabral et al., 2015; Celik et al., 2018; Kaya Kara et al., 2020).
Early Spontaneous Movements in Preterm Infants
Nervous system abnormalities in preterm infants hinder the development of variability and complexity in general movements (Peyton & Einspieler, 2018). The central pattern generator that produces these movements is modulated by supraspinal projections and sensory feedback (Peyton & Einspieler, 2018). Researchers have reported rates of normal fidgety movements in preterm infants of 79% at ≤31 wk gestation (Peyton et al., 2016) and 77% at ≤34 wk gestation (Zang et al., 2016). In studies with extremely preterm infants, researchers found rates of normal fidgety movements of 81% (Örtqvist et al., 2021), 78% (Fjørtoft et al., 2016), and 87% (Sharp et al., 2018); a study with late preterm infants found a rate of normal fidgety movements of 93% (Brogna et al., 2013).
In our study, the rate of normal fidgety movements was 72% in infants born at <32 wk gestation, significantly lower than in infants born at 32 to 36 wk gestation (98%). The rate of normal fidgety movements in the infants born at 32 to 36 wk gestation was higher in our study than in Brogna et al.’s (2013) study with late preterm infants (i.e., 93%). The reason for these different rates may be the variety of other risk factors besides prematurity that were present in the samples.
Age-specific motor repertoire and postural patterns together with fidgety movements have been found to help predict developmental abnormalities, including fine and gross motor performance (Zang et al., 2016), language development (Salavati et al., 2017), and minor neurological dysfunctions (Bruggink et al., 2008). In our study, the infants born at <32 wk gestation had more abnormal movement patterns (especially kicking and mouth movements) and lower MOS results than those born at 32 to 36 wk gestation. Two previous studies comparing MOS results for infants born extremely preterm and term-born control infants found that extremely preterm infants had lower MOS results (Fjørtoft et al., 2016; Örtqvist et al., 2021). Thus, our findings and those of similar studies indicate that developmental abnormalities increase as gestational weeks decrease.
Sensory Processing in Preterm Infants
Infants born at term adapt to a new environment, feeding schedule, and day–night rhythm during the first 2 mo after birth (Fagard et al., 2018). Although they have periods of being awake and interacting with people during this period, most of their time is divided among feeding, crying, and sleeping (Fagard et al., 2018). At ages 2 to 3 mo, they typically begin to discover their own bodies (Fagard et al., 2018). Preterm infants, however, have a different adaptation process, and most stay in the NICU, which, as White-Traut et al. (1994) reported, is not conducive to their sensory development. Sensory processing difficulties may be higher in preterm infants because the NICU environment alters the adaptation process in their first months of life.
Previous studies have reported rates of sensory processing difficulties of 82% in infants at gestation ≤30 wk (Chorna et al., 2014), 60% (Celik et al., 2018) and 73% (Cabral et al., 2015) in infants born at <37 wk gestation, and 47% in late preterm infants (Bart et al., 2011). We observed lower rates of sensory processing difficulties in infants born at 32 to 36 wk gestation than in those born at <32 wk gestation. However, both of our infant groups had lower rates of sensory processing difficulties than found in previous studies (Bart et al., 2011; Cabral et al., 2015; Celik et al., 2018; Chorna et al., 2014). This difference may be attributable to variations in the clinical characteristics (as in early spontaneous movements) or age of the preterm infants studied. In addition, differences between our preterm infant groups may have resulted from variations in risk factors and length of stay in the hospital.
Relationship Between Early Spontaneous Movements and Sensory Processing
Kaya Kara et al. (2020) showed a relationship between motor development and sensory processing in infants born at ≤30 wk gestation in the first 4 mo of life. Celik et al. (2018) reported similar findings in infants born at <37 gestational wk between ages 10 and 12 mo. In contrast, other studies of preterm infants ages 4 to 6 mo (Cabral et al., 2015) and 10 to 15 mo (Case-Smith et al., 1998) found no significant relationship between motor development and sensory processing. These findings also may have been affected by the variety of assessment methods used and differences in the infants’ clinical characteristics. Our results showed no relationship between early spontaneous movements, including fidgety movements, and sensory processing.
Interestingly, Fagard et al. (2018) posited that spontaneous movements cause sensorimotor mapping to emerge. Furthermore, Piontelli (2010) described general movements as sensorimotor storms in which the tactile, proprioceptive, and vestibular senses engage simultaneously. Our findings in preterm infants at ages 3 to 5 mo may reflect progressively emerging sensorimotor mapping in the first months of life. Therefore, we consider that assessing not only motor development but also sensory processing is necessary to predict developmental abnormalities in early life.
Limitations
The Sensory Profile–2 is a parent-report questionnaire. Although parents’ characteristics might play a role in their answers, Eeles, Spittle, et al. (2013) concluded that a parent questionnaire is time efficient and able to capture infants’ behavior in daily life activities. Data on a few parents’ educational level and employment status were missing, but parental characteristics were similar across groups.
Implications for Occupational Therapy Practice
The findings of our study have the following implications for occupational therapy practice: ▪ The preterm infants born at <32 wk gestation had lower scores on early spontaneous movements, including fidgety movements (i.e., lower MOS on the General Movements Assessment), than those born at 32 to 36 wk gestation. Occupational therapy practitioners should be aware that very preterm infants are more likely to have neurodevelopmental problems than moderate– late preterm infants. ▪ The preterm infants born at <32 wk gestation showed more sensory processing difficulties than those born at 32 to 36 wk gestation. Practitioners should advocate for very preterm infants to be evaluated for sensory processing difficulties later in childhood. ▪ We found no relationship between early spontaneous movements and sensory processing in preterm infants. Practitioners should provide assessment and follow-up in both developmental areas for preterm infants in early life. Early evaluation of sensory processing can help practitioners identify signs of sensory processing difficulties and contribute to early intervention with preterm infants.
Conclusion
Both the amount of aberrant fidgety movements and movement patterns and the rate of sensory processing difficulties were higher in preterm infants born at <32 wk gestation than in those born at 32 to 36 wk gestation. Early spontaneous movements and sensory processing may not be related, however, perhaps because of the progressive development of sensorimotor mapping. By assessing both motor development and sensory processing, however, occupational therapy practitioners can play a crucial role in early intervention programs for preterm infants early in life. Further studies are needed to investigate whether or not early spontaneous movements and morbidities in the first months of life have an effect on sensory development at a later age.
Footnotes
Acknowledgments
The authors have no interests that might be perceived as representing any conflict or bias.