Motor coordination difficulties and physical fitness of extremely-low-birthweight children
*Correspondence to second author at Growth and Development Unit, Mater Mothers’ Hospital, Raymond Terrace, South Brisbane 4101, Queensland, Australia. E-mail: firstname.lastname@example.org
Motor coordination difficulties and poor fitness exist in the extremely low birthweight (ELBW) population. This study investigated the relative impact of motor coordination on the fitness of ELBW children aged 11 to 13 years. One hundred and nine children were recruited to the study: 54 ELBW participants (mean age at assessment 12y 6mo; 31 male, 23 female; mean birthweight 769g, SD 148g; mean gestational age 26.6 weeks, SD 2.1 weeks) and 55 comparison children (mean age at assessment 12y 5mo; 28 males, 27 females; at least 37 weeks’ gestation). All children completed the Movement Assessment Battery for Children (MABC), functional tests of postural stability and strength, growth measures, and tests of respiratory function. Maximal oxygen uptake (VO2max) was calculated from a 20m shuttle run test as a measure of fitness. The ELBW group had greater problems with postural stability (p=0.001) and motor coordination (p=0.001), with 70% rated as having a definite motor problem on the MABC brackets (those who scored less than the 5th centile on the MABC). The ELBW was also less fit than the comparison group (p=0.001), with 45% below the 10th centile for VO2max. There were differences between the groups for growth, strength, and particularly respiratory function. However, respiratory function did not significantly correlate with VO2max in the ELBW group. Motor coordination was the most powerful predictor of VO2max in both the ELBW (p=0.001) and the comparison groups (p=0.001).
List of abbreviations
Body mass index
Extremely low birthweight
Forced vital capacity
Forced expiratory volume in the first second of expiration
Movement Assessment Battery for Children
Peak expiratory flow rate
Single leg stance
Maximal oxygen uptake
A high prevalence of motor impairment exists in preschool children who were born very preterm but who do not have a diagnosed disability.1 Motor deficits evident in this group at preschool age including poor coordination and deficits in postural stability, persist through primary school.2–4 Rather than seeing these motor difficulties resolve as the child matures, studies that have followed their cohort over time report an increase in motor problems with age as motor deficits become more apparent with more demanding tasks.5–7 The literature on motor outcomes for extremely low birthweight (ELBW) survivors in adolescence generally reports levels of disability.1,8 However, reports of competency in non-disabled ELBW preterm adolescents, predominately reports from parents and self-report data,9 suggest persistent motor deficits; and studies examining the motor competence of non-disabled low-birthweight children in general do demonstrate persistence of motor deficits into adolescence.10,11
Although poor physical fitness, which has the potential to affect health, learning, and behaviour, is increasingly reported in the literature on ELBW,9,12,13 there is no report about the possible impact motor deficits may have on the physical fitness of these children. As the reported deficits in exercise tolerance in these children have been demonstrated to be independent of their growth14,15 and chronic lung-disease status,12,13 further investigation was needed into the relative impact on fitness of deficits in motor coordination and postural stability.
As fitness levels are reflective both of health outcomes and the competency of the child to participate in activities that promote growth and learning, the emerging evidence of fitness deficits in the extremely preterm population is of concern. This study investigated further the fitness levels and motor competency of non-disabled ELBW children as they were reaching adolescence. It also aimed to determine whether a relationship exists between their motor competence and physical fitness independent of their growth and respiratory status. Additionally, the impact of postural stability deficits on motor coordination and fitness levels was investigated.
The participants of this case–control study were born between January 1992 and December 1994. They were less than 1000g at birth and had been managed in the Neonatal Intensive Care Unit at the Mater Mothers’ Hospital, Brisbane, Australia. Only children who lived within 250km of the testing centre were invited to participate. Children whose motor function was not at least equivalent to Level I of the Gross Motor Function Classification System (GMFCS)16 were excluded. Children were included if they were able to perform gross motor skills including running and jumping, although speed, balance, and coordination may have been reduced.
Of the 113 surviving children who met inclusion criteria, 54 (31 males, 23 females) completed testing. The 59 children who did not participate in the study were either lost to follow-up or unavailable to participate when invited. Table I compares the biological and social characteristics of the group tested with those of the cohort who were not. There was no significant difference between the groups on sex, gestational age, home on oxygen status, or the prevalence of intracranial events including cerebrovascular haemorrhage, ventricular dilatation, or periventricular leukomalacia. The group tested differed from the remainder of the cohort in having a lower mean birthweight and older mothers.
Table I. Comparison of study children and eligible children not included in the study
|Mean birthweight (SD), g||771.1 (147.8)||833.8 (112.3)||0.01|
|Mean gestation (SD), wks||26.5 (2.04)||27.1 (1.96)||0.1|
|Sex, male, %||31 (57.4)||24 (40.7)||0.06|
|Home on oxygen, %||13 (24.1)||15 (25.4)||0.5|
|CVH at least grade 2, %||4 (7.4)||4 (6.8)||0.6|
|Ventricular dilatation, %||7 (13)||9 (15.3)||0.4|
|Cystic PVL, %|
|Mean maternal age (SD), y||31.7 (5.2)||27.25 (5.96)||0.001|
|Maternal education, beyond year 12 completed, %||25 (46.3)||17 (28.8)||0.06|
Fifty-five comparison children (28 males, 27 females), who were term born and of similar age to the preterm group, also completed testing. Comparison children were recruited as a term-born peer who was a school friend of the preterm child or, when that was impractical, a term-born child was recruited from the general community. All children who completed testing were between 11 and 13 years of age and otherwise healthy. Four children with mild cerebral palsy (CP; GMFCS Level I) were included.
The Movement Assessment Battery for Children (MABC) was used to rate motor coordination. The MABC measures manual dexterity, ball skills, and static and dynamic balance. Scores rank the child on centile norms. Results were grouped as described in the MABC manual: as a normal motor skills rating if greater than the 15th centile; as suspect motor competency if between the 15th and 5th centiles; or as a definite motor problem if less than the 5th centile.17 Validity and reliability of the MABC is reportedly high17,18 and has been demonstrated particularly to differentiate motor competence in older children in a study of 8- to 17-year-olds.19
Postural stability was examined using single leg stance (SLS), a functional measure of stability.20 Functional strength was measured with the seated throw test for upper limb strength, and vertical jump for lower limb strength. These functional strength tests are normed for age.21
Height, weight, and head circumference were measured and a body mass index (BMI)22 was calculated for each child.
For respiratory function, a computerized spirometric system, Spirobank (Medical International Research ISO 9001, EN 46001; Rome, Italy), was used to measure forced vital capacity (FVC), forced expiratory volume in the first second of expiration (FEV1), peak expiratory flow rate (PEF), and the ratio of FEV1/FVC. Obstructive airways disease was defined as FEV1/FVC less than 80%.
Cardiorespiratory endurance was measured as an indication of the child’s overall fitness. VO2max was used as the cardiorespiratory endurance measure as calculated from a 20m Shuttle Run test23 using the conversion for children as defined by Leger et al.24 These values are also normed for age. Validity and reliability of the 20m Shuttle Run test as a measure of VO2max was initially established in a study of 188 6- to 16-year-old children.24 Furthermore, the validity of the test has been extended to children with respiratory function impairment in a study of 12- to 17-year-old children with asthma.25
The Mater Health Services Human Research Ethics Committee and The Medical Research Ethics Committee of the University of Queensland granted approval for the study. Written informed consent was obtained from all children and parents. A team of assessors, who were not involved in the follow-up care of any of the participants, was trained in a set protocol of assessment. Those assessing the children received no information about the birth age, birthweight, or background information about the children.
All statistical analysis used SPSS 14.0. All variables were examined to determine their distribution. To examine differences between the groups, independent t-tests were used for continuous normally distributed data, and the Mann–Whitney U test for the nonparametric continuous data of SLS. For categorical measures, cross tabulation and χ2 were used with odds ratio to demonstrate the strength of the association, and the 95% confidence interval (CI) to measure precision. As conversion from Shuttle Run performance to VO2max is dependent on age, data on VO2max were adjusted for age with linear regression. An arbitrary level of 5% statistical significance was assumed (two-tailed).
Correlations for parametric data were examined using Pearson’s coefficient and Spearman’s coefficient for the nonparametric data. Multiple least-squares regression analysis established which of the most powerful motor, growth, and respiratory predictors were independent predictors of VO2max. To establish the relative power of the predictors, the standardized coefficients were also examined. Sex and age at assessment were also included in the regression model. A sensitivity analysis was performed excluding the four children with mild CP to examine the impact on the results.
The measures of respiratory function for four comparison children and one extremely preterm child are missing because of equipment failure. When comparing the groups for differences on respiratory function, we used casewise deletion to deal with the missing data. Regression analysis was restricted to those participants with full data sets (ELBW=53, comparisons children=51).
Data on 54 ELBW children and 55 comparison children were collected on all measures except for respiratory function. For the ELBW group, the mean birthweight was 769g (SD 148g) and the mean gestation was 26.6 weeks (SD 2.1wks). All comparison children were of at least 37 weeks’ gestation. The mean age of the ELBW group was 12 years 6 months (SD 8mo); for the comparison group, the mean age was 12 years 5 months (SD 11mo). There was no significant difference between the groups on age at assessment (p=0.65) or sex mix (p=0.63).
There was a significant difference between the ELBW and the comparison groups in their motor competency as measured by the MABC (p=0.001), with 72.2% of the ELBW group graded as having a definite motor problem compared with 21.8% of the comparison group (Table II). The functional stability measure of SLS (best performance: either right or left leg stand) was markedly superior in the comparison group (difference between medians 214s, p=0.001).
Table II. Movement Assessment Battery for Children (MABC) motor categories in the extremely low birthweight and comparison groups
|Normal motor competency, >15th centile||11 (20.4)||32 (58.2)||1|
|Suspect motor competency, 15th–5th centile||4 (7.4)||11 (20)||1.06 (0.3–4)|
|Definite motor problem, <5th centile||39 (72.2)||12 (21.8)||9.46 (3.7–24.3)|
Table III reports significant differences between the ELBW group and the comparison group on strength, growth, respiratory function, and cardiorespiratory endurance. There was no significant difference between the groups on FVC (p=0.408) but a significant difference between the groups on all other respiratory measures was found: FEV1 (p=0.001), PEF (p=0.002), and FEV1/FVC (p=0.001). When the FEV1/FVC results were grouped as being above or below 80%, the difference between the groups was marked (p=0.001, χ2 13.9, odds ratio 19.74, 95% CI 2.5–156). Of the ELBW group, 28.3% had evidence of obstructive airways disease compared with 2% in the comparison group.
Table III. Comparison of strength, growth, respiratory function, and cardiorespiratory endurance in the extremely low birthweight and comparison groups
|Strength||Seated throw, cm||391.78 (89.64)||453.15 (123.96)||61.4 (20.2, 102.5)||0.004|
|Vertical jump, cm||22.67 (5.15)||25.09 (6.02)||2.42 (0.3, 4.6)||0.026|
|Growth||Height, cm||150.28 (8.53)||155.06 (10.72)||4.8 (1.1, 8.5)||0.012|
|Weight, kg||41.8 (10.87)||48.5 (12.05)||6.7 (2.3, 11.05)||0.003|
|BMI, kg/cm2||18.2 (3.91)||20.05 (3.96)||1.85 (0.35, 3.3)||0.016|
|Head circumference, cm||53.5 (2.13)||55.43 (1.39)||1.93 (1.25, 2.6)||0.001|
|Respiratory functiona||FVC, l||96.96 (12.48)||98.88 (11.02)||1.92 (-2.7, 6.5)||0.4|
|FEV1, l||88.98 (13.47)||97.73 (10.89)||8.75 (3.97, 13.5)||0.001|
|PEF, l/s||75.49 (18.09)||85.47 (14.34)||9.98 (3.6, 16.3)||0.002|
|FEV1/FVC, %||93.26 (7.84)||101.55 (6.05)||8.3 (5.6, 11.02)||0.001|
|Cardiorespiratory endurance||VO2max, ml/kg/min||42.08 (4.9)||46.13 (5.76)||4.05 (2.01, 6.1)||0.001|
Looking at the fitness of the groups, we found the ELBW children did indeed have significantly poorer cardiorespiratory endurance than the comparison group when we compared the means of VO2max (p=0.001). Of the ELBW group, 31.5% were at or below the 1st centile on aged norms for VO2max21 compared with 16.4% of the comparison group; and 44.5% of the ELBW group was at or below the 10th centile for VO2max compared with 18.2% of the comparison group. When VO2max data was adjusted for age with linear regression, the results were unchanged.
The sensitivity analysis excluding the four children with mild CP showed minimal differences only in the analysis of the complete data set. The differences between the groups persisted in MABC rating (χ2 26.8, p=0.001, odds ratio 9.52, 95% CI 3.63–24.9 for those less than the 5th centile versus the normal group), SLS test, VO2max, strength, and measures of respiratory function to similar levels of significance.
To examine the strength of association between VO2max and the motor, growth, and respiratory predictors, the respective correlations were analyzed (Table IV). The strongest correlation for cardiorespiratory endurance across both groups of children was the motor coordination score, with a correlation of 0.5 (p=0.001) for the ELBW group and an even stronger correlation for the comparison group (ρ=0.65, p=0.001). The functional stability measure of SLS and the balance subsection of the MABC score were also significantly correlated with VO2max. Correlation between respiratory function and cardiorespiratory endurance was evident only in the comparison group, the strongest of those correlations being the association of PEF and VO2 max.
Table IV. Correlations of VO2max with motor control, functional stability, strength, growth, and measures of respiratory function
|Motor coordination||MABC total scorea||–0.5 (–0.67to–0.26)||<0.0001||–0.65 (–0.77 to –0.45)||<0.001|
|Stability measures||MABC balance scorea||–0.43 (–0.62 to –0.18)||0.001||–0.52 (–0.69 to –0.29)||<0.001|
|SLS best performancea||0.34 (0.08 to 0.56)||0.012||0.5 (0.27 to 0.67)||<0.001|
|Strength||Seated throwb||0.222 (–0.05 to 0.46)||0.107||0.43 (0.18 to 0.62)||0.001|
|Vertical jumpb||0.34 (0.08 to 0.56)||0.011||0.59 (0.38 to 0.74)||<0.001|
|Growth||Heightb||0.018 (–0.25 to 0.28)||0.896||0.28(0.009 to 0.5)||0.041|
|Weightb||–0.3 (–0.52 to –0.03)||0.030||–0.109 (–0.36 to 0.16)||0.430|
|BMIb||0.32 (–0.54to–0.05)||0.020||–0.34 (–0.55 to –0.08)||0.010|
|Head circumferenceb||–0.139 (–0.39 to 0.135)||0.316||0.148 (–0.12 to 0.4)||0.280|
|Respiratory functionc||FVCb||0.211 (–0.06 to 0.45)||0.129||–0.082 (–0.35 to 0.2)||0.568|
|FEV1b||0.21 (–0.066 to 0.45)||0.131||0.081 (–0.2 to 0.35)||0.571|
|PEFb||0.109 (–0.17 to 0.37)||0.439||0.305 (0.03 to 0.53)||0.029|
|FEV1/FVCb||0.027 (–0.245 to 0.29)||0.849||0.062 (–0.22 to 0.33)||0.666|
It is also noteworthy that the functional stability measure of SLS, although not included in the MABC balance tests, correlated significantly with the MABC balance subsection score in both the ELBW (ρ=0.6, p=0.001) and the comparison groups (ρ=0.33, p=0.001), but that the correlation of SLS with the total motor control score was stronger, with a correlation of ρ=0.66 (p=0.001) for the ELBW group and ρ=0.49 (p=0.001) for the comparison group.
To determine the extent to which motor coordination, growth, and respiratory measures were independent predictors of VO2max, multiple linear regression was used. The motor, growth, and respiratory predictors with the strongest correlation with VO2max were simultaneously entered into the model, with VO2max as the dependent variable. These regression models including MABC, BMI, and FEV1 as predictors were performed separately for the ELBW and comparison groups. Unadjusted and adjusted coefficients are shown in Table V.
Table V. Unadjusted and adjusted regression analysis of the strongest motor, growth, and respiratory predictors of VO2max in extremely low-birthweight and comparison groups
|MABC (95% CI for β)||–0.245 (–0.38 to –0.11)||0.001||–0.2 (–0.34 to –0.07)||0.004|
|BMI (95% CI for β)||–0.4 (–0.73 to –0.07)||0.02||–0.29 (–0.6 to 0.025)||0.07|
|FEV1 (95% CI for β)||0.08 (–0.024 to 0.18)||0.13||0.045 (–0.047 to 0.14)||0.33|
|MABC (95% CI for β)||–0.442 (–0.573 to –0.271)||0.001||–0.386 (–0.56 to –0.212)||0.001|
|BMI (95% CI for β)||–0.498 (–0.874 to –0.122)||0.01||–0.278 (–0.65 to 0.091)||0.14|
|FEV1 (95% CI for β)||0.044 (–0.112 to 0.2)||0.6||0.015 (–0.116 to 0.145)||0.8|
In the adjusted analysis, only the MABC independently predicts VO2max in either the ELBW or the comparison groups. For the ELBW group, the standardized adjusted coefficients on MABC, BMI, and FEV1 scores were −0.385 (p=0.004), −0.230 (p=0.07), and 0.12 (p=0.3) respectively; for the comparison group they were –0.55 (p=0.001), –0.19 (p=0.14), and 0.03 (p=0.8) respectively. Addition of age and sex to these models did not substantially alter the findings. The sensitivity analysis excluding the children with CP from the multiple regression produced similar results. For the MABC, the adjusted values for the ELBW group were β=−0.193 (95% CI −0.338 to −0.049), and for the comparison group β=−0.429 (95% CI −0.628 to −0.23).
The present study has found a high prevalence of definite motor coordination problems in the cohort of non-disabled ELBW children at 11 to 13 years of age, as defined by results on the MABC below the 5th centile. A significant fitness deficit in this cohort was also demonstrated, with almost half of the ELBW group at or below the 10th centile for VO2max for their age. Despite a high prevalence of respiratory function indicative of obstructive airways disease in the ELBW children (28.3%), there was no significant association between respiratory function and the fitness measure of VO2max in the ELBW group. Motor coordination was the only measure that proved to be an independent predictor of fitness in both the ELBW and comparison groups. Functional stability proved to be strongly associated with motor coordination in both groups.
The prevalence of the impairment of motor coordination in this cohort of ELBW children with no overt disability was greater than the prevalence reported in younger cohorts of preterm children born.2,5 These findings extend the trend of increasing prevalence of evident motor difficulties in the non-disabled ELBW population with age that has been demonstrated in follow-up studies of preschool children.6,7 A recent study11 of non-disabled low birthweight survivors through to 16 years of age indicated that motor deficits in this population of 11- to 13-year-old extremely preterm survivors are likely to persist.
Significant problems of poor physical fitness in this extremely preterm group confirm other findings of increased fatigability13and poor fitness14 in younger cohorts. The findings of Rogers et al.9 of persistent fitness deficits in the 17-year-old unimpaired ELBW children suggest that these children will not ‘catch up’ or grow out of their limitations in fitness.
Previous studies have implicated motor coordination as a factor impeding fitness in very low birthweight populations, as the deficits in aerobic and anaerobic performance could not be accounted for by body size,14 muscle mass,15 or pulmonary function status.12 This study did find deficits in growth, strength, and particularly respiratory function in the extremely preterm group compared with the comparison group. However, as in those earlier studies, we demonstrated only a slight association with growth and strength measures and the ELBW children’s fitness measure of VO2max. Although there was mild association of respiratory function deficit and fitness in the comparison group, in the ELBW group, of whom 28.3% had obstructive airways disease, there was no significant correlation between VO2max and any measure of respiratory function.
This study did, however, find VO2max strongly correlated with motor coordination in both the ELBW and comparison groups. Motor coordination proved to be the only independent predictor of VO2max for both groups.
Just as motor control proved to be the strong correlate for VO2max, the functional stability measure of SLS strongly correlated with the motor control score. This reflects what is observed clinically that, in children of very preterm birth, motor problems are often associated with poor postural stability.4,26,27 It is noteworthy that the correlation between motor competency and postural stability is also marked in the comparison group. This finding supports other studies that have examined the relationship of postural control and motor coordination in children of normal birthweight.28,29
In the context that postural stability has been demonstrated to be important in the development of motor coordination, the strong correlation of SLS with the MABC total and with the MABC balance subscore recommends SLS as a useful and easy functional test of stability when examining children at risk of poor motor coordination or fitness.
The inclusion criteria of motor competence of at least GMFCS level I allowed the study to investigate the outcome for ELBW children who have minimal or mild motor problems regardless of the underlying aetiology of those motor deficits. The sensitivity analysis showed minimal differences to the outcomes only when children with CP were excluded.
Although assessors were blind to the birth status of each child tested, growth deficits made it difficult to be truly blind to the status of some of the preterm group. However, all the tests were categorical, with little subjectivity in grading performance. The prevalence of motor impairment in the comparison group also contributed to confounding testers, as only 58.2% of that group scored above the 15th centile for motor competency on the MABC. This may be due to the method of recruitment, whereby the children of very preterm birth were asked to bring a friend. The less than average motor competency of the comparison group, however, acts to minimize the difference between the ELBW group and the normal birthweight children, a difference that was still demonstrated to be highly significant. Using normed tests allowed the issue of motor competence of the comparison group to be noted and for the scale of the deficit in performance of the ELBW group to be demonstrated.
The ELBW group tested was restricted to that portion of the complete cohort that could attend the testing centre. However, the tested and untested ELBW children were not significantly different for perinatal variables, except for the former having smaller birthweight and older mothers. Socioeconomic factors were not examined in this study. However, as socioeconomic status has not been shown to predict motor performance independently,11 the effect of motor coordination on fitness demonstrated in this group would be expected to hold for children of this age group generally.
Fitness is primarily dependent on motor competence rather than the degree of impairment in respiratory function in 11- to 13-year-old ELBW children. Motor coordination is also the principal determinant of cardiorespiratory endurance in the comparison group. Deficits in motor coordination and fitness in the ELBW group were marked, with 70% of the children born extremely preterm but who were otherwise unimpaired rating as having a definite motor problem, and 45% of the group falling below the 10th centile for their cardiorespiratory endurance. The functional stability measure of SLS proved to be strongly correlated with motor control and so may be clinically useful in screening children.
This study indicates a need to support ELBW children in the development of their motor competence as a significant determinant of their general fitness. The information allows strategies to be planned to support this group. It can also be used to educate families about the increased prevalence of motor coordination difficulties in ELBW children and that these problems can disadvantage them in their functioning and fitness. Demonstrating the association of postural stability, motor coordination, and fitness helps us to support these children more effectively and to deliver services to them more efficiently.
We thank the Mater Foundation for providing financial support for the study.