Kinematic analysis of kicking movements in preterm infants with very low birth weight and full-term infants. (Research Report).

Advancements in perinatal care in past decades have resulted in a marked improvement in the survival rate of preterm infants with very low birth weight (VLBW) (birth weight of <1,501 g and gestational age [GA] of <37 weeks). (1) Surviving infants, however, sustain an increased risk of long-term developmental disabilities, with approximately 5% to 15% exhibiting major neurodevelopmental disabilities (eg, cerebral palsy) and with an additional 25% to 50% exhibiting mild to moderate neurodevelopmental disabilities. (2,3) Current approaches to management of preterm infants with neuromotor disorders have emphasized early assessment and intervention. (4,5)

Spontaneous motility constitutes the major part of the motor repertoire of infants during their early months of life. (6) These self-initiated motor behaviors involve movements of various body parts (upper extremities, lower extremities, neck and trunk) with a fluent and complex appearance, and they are variable in speed, amplitude, and intensity. Recent evidence indicates that newborn infants sustaining adverse events such as apnea, (7) chronic lung disease, (8) and brain damage (9,10) during the neonatal period tend to show alterations in the quality components of spontaneous movements in infants. Furthermore, the presence of such abnormalities in early infancy has been shown to correlate with poor neurologic outcome such as cerebral palsy. (8,11,12) These results suggest that assessment of spontaneous motility may be a useful and sensitive method of identifying early neuromotor problems in preterm infants.

Among early motor behaviors, kicking movements have aroused great interest because of their potential role in the development of locomotion. (13) Thelen and colleagues (13-17) conducted a series of follow-up studies on full-term infants to examine their early kicking movements and relate those movements to the walking ability they later developed. Infants exhibited patterns of kicking movements during their newborn period that have high interjoint coordination (indicated by high pair-wise cross-correlations of joint angles at the hip, knee, and ankle), small phase lags (indicated by small differences between the times at which the joints started moving or reached peak excursion), and constrained movement durations (indicated by invariant durations in the flexion and extension phases). (13,17) With maturation, kicking behavior showed reorganization from the newborn stereotyped movements toward more skilled and complex activities. Nevertheless, the spatial and temporal patterns of early kicking movements were found to be similar to those involved in mature walking cycles. (13,17) The results indicate that kicking may be a developmental manifestation of a central program later used for locomotion. Therefore, investigation of kicking development in preterm infants may provide a unique means to understand how the earliest spontaneous movements change in these children as they acquire voluntary control.

Assessment of kicking has often been aided with kinematic analysis to measure qualitative aspects of movement. (18) Kinematic analysis often involves imaging techniques (eg, television, cinematography, optoelectric systems) that are used to identify particular points on the body and to track these points in space over time. (19) Imaging techniques can provide a quantitative assessment of movement that allows us to transform a visually observed movement into something measurable. (20) Furthermore, the techniques can provide a comprehensive description of movement patterns of infants by delineating time, displacement, velocity and acceleration, and movement coordination among the body segments. (19) Several researchers have investigated the early kicking movements of preterm infants using kinematic analysis, (9,10,21-25) however, the movement patterns they described were inconsistent (Tab. 1). Kinematic variables included kick frequency, spatiotemporal organization, interjoint coordination, and interlimb coordination. For example, van der Heide et al (10) reported that preterm infants who were at low risk for neurodevelopmental disabilities exhibited a lower correlation of the hip-knee and hip-ankle couplings, indicating low interjoint coordination, than preterm infants with brain damage and full-term infants at 1 month of corrected age (adjusted for prematurity) and that the differences resolved at older ages. In contrast, Vaal et al (25) found that full-term infants and preterm infants at low risk for neurodevelopmental disabilities had a lower correlation of the hip-knee, hip-ankle, and knee-ankle couplings than preterm infants with brain damage at 18 and 26 weeks of corrected age. Droit et al (23) reported that preterm infants with brain damage showed a higher level of semi-both-leg kicking (simultaneous flexion and nonsimultaneous extension) and less alternate kicking than preterm infants at low risk for neurodevelopmental disabilities during the neonatal period. In contrast, van der Heide et al (10) observed no difference in the interlimb coordination patterns between preterm infants with brain damage and preterm infants at low risk for neurodevelopmental disabilities at 1 and 3 months of corrected age.

The lack of agreement among the results of previous studies might be due to variations in sample characteristics (eg, differences in GA and incidence of medical complications) and methods (eg, differences in study design and definitions of kinematic variables). Therefore, the purpose of our study was to examine the development of kicking movements with kinematic analysis in preterm infants with VLBW and full-term infants from 2 to 4 months of corrected age. The incorporation of kinematic analysis in a longitudinal follow-up design can provide a quantitative picture of kicking movements that can, in theory, help us understand how kicking movements mature during early infancy. The examination of kicking movements was carried out on infants at 2 and 4 months of corrected age because major reorganization in motor function may occur during this age period. (26) Furthermore, the inclusion of a range of kinematic variables (kick frequency, spatiotemporal organization, interjoint coordination, and interlimb coordination) was designed to examine the full range and complexity of kicking development.

Method

Subjects

We recruited preterm infants with VLBW who were admitted to the neonatal intensive care unit at National Taiwan University Hospital, Taipei, Taiwan. The inclusion criteria were: GA of <37 weeks, birth weight of <1,501 g, and absence of congenital or chromosomal abnormality. Chromosomal evaluation is a routinely performed procedure at our hospital when infants show facial or major anomaly. Full-term infants born at the same hospital also were included to serve as a comparison group. The selection criteria for these infants were: GA between 38 and 42 weeks, birth weight within the 10th to 90th percentiles of the intrauterine growth curve of Taiwanese infants, (27) 1- and 5-minute Apgar scores greater than 8, absence of prenatal and perinatal complications, and lack of identifiable pathology or neurological deficits when examined as newborns. Informed parental consent was obtained for all infants prior to participation in the study.

From June 1997 to December 1998, a total of 22 infants with VLBW and 22 full-term infants were included in the study. The groups were matched for levels of maternal education, stratified into greater than high school, high school, and less than high school. The perinatal characteristics (ie, GA, birth weight, 1- and 5-minute Apgar scores, presence of chronic lung disease [CLD], presence of intraventricular hemorrhage [IVH], and presence of retinopathy of prematurity [ROP]) of the groups are presented in Table 2. Gestational age was estimated by maternal menstruation date and confirmed by prenatal ultrasonography. Chronic lung disease was defined as lung disease requiring a dependence on oxygen therapy for a duration of more than 28 days? The severity of

IVH was classified into grades I to IV according to the method of Papile et al. (29) The severity of ROP was graded into stages I to IV according to the International Classification of Retinopathy of Prematurity. (30)

The infants with VLBW were further stratified into low GA (GA of <30 weeks, n=9) and high GA (GA of [greater than or equal to] 30 weeks, n=13) classes. This distinction was made on the basis of findings that a short gestation increases the risk of developmental delays in preterm infants. (31,32) The infants with a low GA had lower birth weights (t[20]=3.84, P=.01), lower 1-minute Apgar scores (t[20]=2.2, P=.04), and higher incidences of CLD ([X.sup.2] [1]=22, P=.001) and ROP ([X.sup.2] [2]=7.1, P=.03) than the infants with a high GA. However, the groups were comparable in the distributions of sex, severity of IVH, and maternal education.

Instrument and Procedures

The infants' kicking movements were recorded, and they were assessed for body weight at 2 and 4 months of corrected age at the neonatal follow-up clinic. During testing, the infants were placed in a supine position on an examination table with a large surface. They wore a specially designed diaper and black shorts to expose the following anatomical landmarks bilaterally: mid-trunk, greater trochanter, lateral femoral epicondyle, lateral malleolus, and fifth metatarsal head (Fig. 1). Reflective ball-shaped markers with a diameter of 1 to 2 cm were placed over these landmarks to define the hip, knee, and ankle joint angles. Infants were tested I hour before feeding and were kept in an alert state (state 5, according to the Brazelton Neonatal Behavioral Assessment Scale (33)). Each infant's head was maintained in the body's midline by an investigator's hand in an effort to control for the effect of tonic neck reflex. Infants were allowed to kick spontaneously for 5 minutes, with the parents or toys in view to minimize frustration.

[FIGURE 1 OMITTED]

The kicking movements of the infants was recorded using 4 synchronized cameras operating at 60 Hz (2 Peak High Speed Video cameras* with lenses of 12.5-75 mm and 2 WV CL-350 video cameras [dagger] with lenses of 8-48 mm). Each camera was connected to a video recorder (SVHS AG1960 and AG1970[dagger]), an SMPTE Time Code Generator (Horita SR-50 [dagger]), an Event & Video Control Unit,* and a monitor [double dagger] (Trinitron [section]). Two cameras were placed on each side of the infants at a distance of 2 m and at the same height as the table surface on which the infants were lying. The cameras were focused on the infants and were angled at 70 degrees to construct a 3-dimensional analysis of the movements of the same limb. The calibration errors, as calculated by the differences between the coordinates of markers measured with cameras and the coordinates of markers in the 3-dimensional space, were of less than 3 mm. (34)

Data Acquisition and Reduction

A 20-second segment of video recording of continuous active kicking movements was selected for each infant. The video data were digitized and analyzed using the Peak Performance Motion Analysis System (Peak Motus Version 3.01* with a fourth-order Butterworth filter and filtering rate of 6 Hz). The cameras essentially provided a 2-dimensional video image of a 3-dimensional scene on a medium. The 2-dimensional video data set generated from the cameras that recorded the same limb movements was converted into 3-dimensional spatial coordinates using the direct linear transformation method. (34) The linear and angular data obtained were used to calculate the variables of kick frequency, spatiotemporal organization, interjoint coordination, and interlimb coordination.

A kick was defined as a flexion and then an extension movement of a lower limb, in which the hip angle change during flexion exceeded 11.5 degrees and the hip angle change during extension was a minimum of 20% of the angular displacement during flexion. (35) The kick cycle was characterized by 4 phases: flexion, intra-kick pause, extension, and inter-kick pause (see Fig. 2 for definitions of phases). (13) The kick frequency was measured as the number of kicks by each leg during the 20-second period and was converted to cycles per minute. Spatio-temporal variables included the kick amplitude and the 4 kick phases. Kick amplitude was defined as the excursion of hip flexion during a kick cycle. Interjoint coordination was measured by pair-wise cross-correlations of the joint angles at the hip, knee, and ankle for the same limb during the kick cycles (excluding the inter-kick pause) using Pearson product moment correlations (r), and these measurements were transformed to Fisher Z scores. (9) Interlimb coordination was measured by frequency (expressed as a percentage) of alternate, unilateral, and synchronous kicking patterns during the 20-second period. (14) An alternate kick was defined as simultaneous flexion of one lower extremity and extension of the other lower extremity, with the flexion phase of the 2 limbs overlapping for less than 50% of the movement. A unilateral kick was defined as isolated flexion and extension of one lower extremity while the other lower extremity was in the intra-kick pause phase or the inter-kick pause phase. A synchronous kick was defined as simultaneous flexion (or extension) of both legs during more than 50% of the flexion (or extension) phase, which was determined by the phase duration of the leg exhibiting a higher kick frequency. The predominant interlimb coordination pattern was also determined for each infant as defined by the most frequent type of kicking pattern that occurred.

[FIGURE 2 OMITTED]

Data Analysis

Kick frequency, spatiotemporal organization, and interjoint coordination were compared between limbs using the paired t test. The collinearity among kicking variables was examined using Pearson product moment correlation coefficients. The kicking variables were compared between groups across ages using an analysis of variance (ANOVA) for repeated measures, in which group was treated as a factor and age was considered as an independent variable. For the variables showing interactions, post hoc Tukey tests were used to examine the level of differences between groups for each age and the level of difference between ages for each group. For the variables showing a group effect but a no interaction, post hoc Tukey tests were used to examine the level of difference between groups. The predominant kicking patterns of the groups were compared across ages using chi-square tests. All statistical analyses were performed with the Statistical Analysis Software (SAS) program (Version 6.2). A probability value of less then .05 was defined as statistically significant. To reduce the probability of type I errors with multiple comparisons, the acceptable level for statistical significance in post hoc tests was lowered to .006 (.05/9).

Results

Comparison of kick frequency, spatiotemporal organization, and interjoint coordination between limbs revealed differences (all P<.05), but there was no preference of limb use. Accordingly, measurements obtained from the kicking performance of 2 limbs were treated as independent data. Furthermore, correlation analysis of all kicking variables showed low to moderate degrees of correlation (all r<.6), indicating minimum collinearity among variables. Therefore, all kicking variables were compared between the infants with VLBW and a low GA, the infants with VLBW and a high GA, and the full-term infants across ages.

Kick Frequency

Analysis of the kick frequency showed no age or group effects but did show an interaction effect (F=10.8; df=2,172; P>.0001) (Tab. 3). Post hoc tests revealed a comparable kick frequency among groups at 2 months of corrected age, but a higher kick frequency in the infants with VLBW and a low GA at 4 months of corrected age (P=.0001). The full-term infants and the infants with VLBW and a high GA exhibited no change in kick frequency across ages, whereas the infants with VLBW and a low GA showed an increase in kick frequency (P<.002).

Spatiotemporal Organization

Age effects (F=60.9; df=1,173; P<.001) and interaction effects (F=3.9; df=2,172; P<.001) were found for the kick amplitude (Tab. 3). Post hoc tests revealed an increase in kick amplitude across ages in all infants. Group effects (F=3.7; df=2,172; P<.05), age effects (F=40.8; df=1,173; P<.001) and interaction effects (F=4.3; df=2,172; P<.05) were obtained for the flexion phase. Post hoc tests revealed a comparable flexion phase among groups at 2 months of corrected age, but a shorter flexion phase in the infants with VLBW and a low GA at 4 months of corrected age (P=.002). The full-term infants and the infants with VLBW and a high GA exhibited a longer flexion phase with age, whereas the infants with VLBW and a low GA manifested a stable flexion phase. Examination of the intra-kick and inter-kick pauses revealed age effects (F=14.6; df=l,173; P<.001 and F=6.5; df=l,158; P<.05) but no group effect. All infants manifested a longer intra-kick pause and a shorter inter-kick pause across ages. As for the extension phase, the performance was comparable among groups and across ages.

Interjoint Coordination

Group effects (F=50.5; df=2,172; P<.001) and age effects (F-3; df=l,173; P>.05) were found for the hip-knee coordination (Fig. 3A). Post hoc tests revealed a borderline higher hip-knee correlation in the infants with VLBW and a low GA than in the full-term infants and the infants with VLBW and a high GA (P=.017 and .048). All groups manifested a decline in the hip-knee correlation with age. The correlation values for the hip-ankle and knee-ankle coordinations were comparable among groups and across ages (Figs. 3B and 3C).

[FIGURE 3A, 3B AND 3C OMITTED]

Intorlimb Coordination

No group or age effects were found for the percentage of alternate kicks (Fig. 4A). However, age effects were obtained for the percentage of unilateral kicks (F=ll.8; df=1,86; P<.001) and synchronous kicks (F=25; df=l,86; P<.001) (Figs. 4B and 4C). All infants exhibited a decreasing percentage of unilateral kicks and an increasing percentage of synchronous kicks across ages. Examination of the predominant kicking pattern revealed different distributions in the infants with VLBW and a low GA across ages and that they frequently kicked with a unilateral pattern at 2 months of corrected age and with a synchronous pattern at 4 months of corrected age ([X.sup.2]=11.8, df=3, P=.003) (Tab. 4). The other groups, however, exhibited no change in the distribution of predominant kicking pattern.

[FIGURE 4A, 4B AND 4C OMITTED]

Correlation analysis of the postnatal body weight with all kicking variables showed a correlation between the body weight and kick amplitude at 2 months of corrected age (r=.38). Because no group effect was found for the kick amplitude in the results of the ANOVA for repeated measures, there was no need to re-examine the effect of kick amplitude in adjusting for body weight.

Discussion

This study examined the kicking movements of preterm infants with VLBW and full-term infants with kinematic analysis at 2 and 4 months of corrected age. The infants with VLBW and a high GA manifested kicking movements and developmental trends similar to those of the full-term infants during the follow-up period. This finding was in contrast to the kicking movements and developmental trends of the infants with VLBW and a low GA, who showed differences in several aspects of kicking performance. The distinct kicking features of the infants with VLBW and a low GA included a higher kick frequency and a shorter flexion phase at 4 months of corrected age and a higher hip-knee correlation and lower variability in interlimb coordination pattern at 2 and 4 months of corrected age. Our findings indicate that infants with VLBW, particularly those with short gestation, have age-related differences in movement organization and coordination of kicking than full-term infants.

We observed a similar tendency for the full-term infants and the infants with VLBW and a high GA to show a decreased kick frequency from 2 to 4 months of corrected age. The infants with VLBW and a low GA began with a lower kick frequency, but subsequently the kick frequency increased to levels higher than those of the full-term infants at 4 months of corrected age. The differences in kick frequency between preterm and full-term infants may relate, in part, to differences in their physical condition and neurological development.

Physical condition could affect a child's physiological energy and activity level. (9) In our study, a low GA was associated with less-optimal conditions (ie, low Apgar scores and a high incidence of CLD), which might initially limit an infant's physiological energy and therefore the infant's ability to sustain high activity levels. (22,36,37) The gradual recovery of their physical condition from the perinatal insults as time passes, in our view, should improve the child's activity levels and thus increase the child's kicking movements. Furthermore, development of the nervous system may also influence the motor behavior of infants. (38) van Wulfften and Hopkins (39) documented a broad spectrum of interrelated changes in neurological functions (eg, learning abilities, motor behavior, social competence) in full-term infants from 2 to 4 months of age. Hopkins and Precht138 observed that motor behavior specifically shows reorganization from newborn spontaneous movement to achievement of fine-distal, goal-directed movement. The full-term infants and the infants with VLBW and a high GA in our study may have undergone age-appropriate changes in neural functions so that toward 4 months of corrected age, they exhibited only a few kicks and began to play more with their hands, reach for toys, smile, and vocalize.

We found a similar spatiotemporal organization of kicking among the groups, except for the flexion phase. The flexion phase durations were comparable among infants at 2 months of corrected age. Toward 4 months of corrected age, the flexion phase durations were extended in the full-term infants and the infants with VLBW and a high GA but remained unchanged in the infants with VLBW and a low GA. The findings contrasted with those of previous reports (9,10) in which the flexion phase durations were consistent in preterm and full-term infants across ages (around 0.4 second). This discrepancy may be accounted for by the use of different joint excursions to determine the presence of kicking. Geerdink et al (9) and van der Heide et al (10) determined the flexion and extension movement of kicking based on knee joint excursions, (17) whereas we studied hip joint excursions based on a previous report by Jensen et al. (35) Examination of the kicking movements of the individual infants at 2 months of corrected age revealed rapid and nearly simultaneous flexion (or extension) in the hip and knee that may explain the equivalent durations (around 0.4 second) obtained with the 2 methods. The joint action became more individualized with maturation in the full-term infants and the infants with VLBW and a high GA as their flexions at 4 months of corrected age consisted of an initial flexion of the hip and knee to a semiflexed position, followed by a continuous flexion in the hip with the knee in extension. The inclusion of the time period that the knee is dissociated from the synergistic patterns may have resulted in a longer flexion phase duration during the hip joint excursions.

We identified hip-knee coordination as the only interjoint coupling that distinguished the kicking movements between groups. Despite a decline in the correlation values from 2 to 4 months of corrected age, the hip-knee coordination of the infants with VLBW and a low GA remained consistently higher than those of the full-term infants and the infants with VLBW and a high GA throughout the period. A study of kicking development in full-term infants by Thelen (15) also showed high pairwise joint correlations among the hip, knee, and ankle at a newborn age, followed by a decrease in the joint correlations. The earliest high interjoint correlations may be the result of a very simple coactivation of the 3 joints. The subsequent increase of joint independence, as denoted by declined joint correlation, may provide the variability of movement that allows the emergence of new movement pattern. (15) Studies are needed to examine whether the high hip-knee correlation in infants with VLBW and a low GA would limit their ability to acquire selective controls and adaptive motor function at later ages.

Our results failed to support the discriminative usefulness of hip-ankle and knee-ankle coordination that has been reported. (9,10,25) This may have been due to differences in sample characteristics and instrumentation between our study and the previous studies. (9,10,25) Unlike the studies by Geerdink et al, (9) van der Heide et al, (10) and Vaal et al (25) where selected preterm infants with severe brain damage or Without major complications were chosen for comparison, in our study we included preterm infants with VLBW and complications of CLD or ROP. Future investigation is necessary to examine the effects of specific neonatal diseases on the kicking movements of preterm infants. In addition, differences in instrumentation may influence the joint motion data obtained from kinematic analysis. Previous researchers (9,10) used one camera for 2-dimensional kinematic analysis to assess single-leg movements. This method contrasts with that of our study, in which we used 4 synchronized cameras to construct a 3-dimensional analysis of bilateral lower-extremity movements. Preterm infants tend to show rotational movements at the hip and ankle during kicking, (22) which are more accurately evaluated by the use of instrumentation to construct a 3-dimensional analysis. (40)

Our results replicated the data of Vaal et al, (25) who demonstrated that preterm and full-term infants exhibited similar developmental trends of interlimb coordination as they changed from frequently kicking with a unilateral pattern at 2 months of corrected age to a synchronous pattern at 4 months of corrected age. According to the observations of Touwen, (41) full-term infants develop spontaneous leg movements, which follows the sequence from neonatal alternate movements to predominantly asymmetrical movements with a single lower extremity at 6 weeks of age, to predominantly symmetrical movements at 3 months of age, and to largely voluntary movements at 5 months of age. Our findings indicate that prematurity does not alter the sequence of developmental changes in bilateral movement coordination of spontaneous lower-extremity movements.

Short gestation appears to influence the variability of interlimb coordination in the development of kicking movements. The full-term infants and the infants with VLBW and a high GA showed large variation in the distribution of kicking patterns. Although most infants adopted unilateral kicks as their predominant pattern at 2 months of corrected age and the synchronous kicks at 4 months of corrected age, some of the infants had assumed other forms of coordination as the predominant pattern. In contrast, the infants with VLBW and a low GA manifested a skewed distribution of kicking patterns. An extremely high proportion adopted the unilateral kicking pattern at 2 months of corrected age, and this was repeated for the synchronous kicking pattern at 4 months of corrected age. Variability in infant behavior has been proposed to have evolutional and functional benefits because it may facilitate the adaptation of an individual child or a population to an ever-changing environment. (42) The skewed distribution of kicking patterns in the infants with VLBW and a low GA, therefore, may imply a lack of variability or flexibility in adjusting bilateral movement coordination to environmental changes during the development of lower-extremity movements.

Conclusion

Our study showed similar kicking movements and developmental trends among the infants with VLBW and a high GA and the full-term infants at 2 and 4 months of corrected age. In contrast, the infants with VLBW and a low GA exhibited different kicking movements, including a higher kick frequency and a shorter flexion phase at 4 months of corrected age and a higher hip-knee correlation and a lower variability in interlimb coordination patterns at 2 and 4 months of corrected age. The findings indicate that infants with VLBW, particularly those with a low GA, have more distinct kicking movements than full-term infants, as characterized by agerelated differences in movement organization and coordination. Our incorporation of kinematic analysis in this longitudinal follow-up study provides a quantitative assessment of qualitative differences in the kicking movements of infants born preterm and at full term. The information that we obtained contributes to an understanding of how spontaneous motor behaviors mature within individuals during early infancy. The relationship between early kicking movements and walking ability is currently under investigation and needs to be established before the initial differences in kicking movements can be used to predict subsequent developmental disabilities in preterm infants.

* Peak Performances Technologies Inc, 7388 S Revere Pkwy, #601, Englewood, CO 80112.

[dagger] Panasonic Broadcast & Television System Co, 1 Panasonic Way, Secaucus, NJ 07094.

[double dagger] Horita Co, PO Box 3993, Mission Viejo, CA 92680.

[section] Sony Corp of America, 1 Sony Dr, Park Ridge, NJ 07656.

* Peak Performances Technologies Inc, 7388 S Revere Pkwy, #601, Englewood, CO 80112.

SAS Institute Inc, PO Box 8000, Cary, NC 27511.

Table 1.
Comparison of Kinematic Studies of Kicking Movements Among Preterm and
Full-Term Infants (a)

Source                                           Study
(Year)                     Samples               Design

Heriza (21,22) (1988)      10 LR preterm         Cross-sectional
                           15 full-term

Droit et al (23) (1996)    6 LR preterm          Cross-sectional
                           6 preterm with BD

Geerdink et al (9)         10 preterm            Longitudinal
  (1996)                   13 full-term            follow-up

van der Heide et al (10)   12 LR preterm         Cross-sectional
  (1999)                   11 preterm with BD
                           12 full-term

Vaal et al (25) (2000)     10 full-term and LR   Longitudinal
                             preterm               follow-up
                           19 preterm with BD

Our study                  22 preterm            Longitudinal
                           22 full-term            follow-up

Source                                     Instrument
(Year)                     Testing Age     (Model)

Heriza (21,22) (1988)      PCA 40 wk       1 camera (2-D)

Droit et al (23) (1996)    PCA 31-35 and   1 camera (2-D)
                             37-39 wk

Geerdink et al (9)         CA 6, 12, and   2 cameras (2-D)
  (1996)                     18 wk

van der Heide et al (10)   CA 1 and 3 mo   1 camera (2-D)
  (1999)

Vaal et al (25) (2000)     CA 6, 12, 18,   4 cameras (3-D)
                             and 26 wk

Our study                  CA 2 and 4 mo   4 cameras (3-D)

Source                     Kinematic Variables
(Year)                     Studied

Heriza (21,22) (1988)      Frequency, (b) phase lag, phases, (b)
                             and amplitude

Droit et al (23) (1996)    Bouts, phases, (b) and interlimb
                             coordination (b)

Geerdink et al (9)         Frequency, (b) phases (b) and
  (1996)                     interjoint coordination (b)

van der Heide et al (10)   Frequency, phases, interjoint
  (1999)                     coodination, (b) and interlimb
                             coordination

Vaal et al (25) (2000)     Phases, variability of
                             amplitude, (b) interjoint
                             coordination, (b) and interlimb
                             coordination

Our study                  Frequency, (b) phases, (b) interjoint
                             coordination, (b) and interlimb
                             coordination (b)

(a) LR=low risk, BD=brain damaged, PCA=post-conceptional age,
CA=corrected age, 2-D=two-dimensional, 3-D=three-dimensional.

(b) Kinematic variables that were shown to have discriminative values.
Table 2.
Perinatal Characteristics of Preterm Infants With Very Low Birth
Weight (VLBW) and Full-Term Infants (a)

                                    Infants With VLBIA
                                    Low GA               High GA
Characteristic                      (n-9)                (n=13)

Male gender, N (%)                       7 (78)              8 (62)
GA (wk)
  X                                    27.7 (b)             31.7b
  SD                                    1.2                  1.7
  Range                                26-29                30-35
Weight (g)
  Birth
    X                                 995 (b)            1,309 (b)
    SD                                237                  148
    Range                             636-1,352            992-1,492

  2 mo of corrected age
    X                               4,306                4,835
    SD                              1,110                  890
    Range                           2,586-6,000          3,628-6,455

  4 mo of corrected age
    X                               5,695                5,945
    SD                              1,244                  742
    Range                           3,800-7,105          4,940-7,080

Apgar score (points)
  1 min
    X                                   4.0 (b)              6.0 (b)
    SD                                  2.2                  2.0
    Range                               1-7                  1-9

  5 min
    X                                   7.1                  8.2
    SD                                  1.8                  1.1
    Range                               4-9                  5-9

Chronic lung disease, N (%)             9 (100) (b)          0 (0) (b)

Intraventricular hemorrhage, N (%)
  Normal                                7 (78)               12 (92)
  Grades I-II                           2 (22)                1 (8)
  Grades III-IV                         0 (0)                 0 (0)

Retinopathy of prematurity, N (%)
  Normal                                5 (56) (b)           13 (100)b
  Stages I-II                           3 (33) (b)            0 (0)b
  Stages Ill-IV                         1 (11) (b)            0 (0)b

                                    Total Sample     Full-Term Infants
Characteristic                      (N=22)           (N=22)

Male gender, N (%)                     15 (68) (c)       8 (36)c
GA (wk)
  X                                    30.1 (c)         39.1 (c)
  SD                                    2.5              3.1
  Range                                26-35            38-41
Weight (g)
  Birth
    X                               1,180 (c)        3,298 (c)
    SD                                243              219
    Range                             636-1,492      2,910-3,632

  2 mo of corrected age
    X                               4,619 (c)        5,436 (c)
    SD                                997              423
    Range                           2,586-6,455      4,800-6,350

  4 mo of corrected age
    X                               5,843 (c)        6,723 (c)
    SD                                960              638
    Range                           3,800-7,105      5,886-7,900

Apgar score (points)
  1 min
    X                                   5.2 (c)          9.0 (c)
    SD                                  2.3              0.0
    Range                               1-9              9-9

  5 min
    X                                   7.7 (c)          9.0 (c)
    SD                                  1.5              0.0
    Range                               4-9              9-9

Chronic lung disease, N (%)             9 (41) (c)       0 (0) (c)

Intraventricular hemorrhage, N (%)
  Normal                               19 (86)
  Grades I-II                           3 (14)
  Grades III-IV                         0 (0)

Retinopathy of prematurity, N (%)
  Normal                               18 (81)
  Stages I-II                           3 (14)
  Stages Ill-IV                         1 (5)

(a) GA= gestational age.

(b) Significant differences (P<.05) were obtained among infants with
VLBW and a low GA and infants with VLBW and a high GA; t tests were
used for continuous variables; chi-square tests were used for
categorical variables.

(c) Significant differences (PC.05) were obtained among infants with
VLBW and full-term infants; t tests were used for continuous variables;
chi-square tests were used for categorical variables.
Table 3.
Kick Frequency and Spatiotemporal Organization of Kicking Movements in
Preterm Infants With Very Low Birth Weight (VLBW) and Full-Term Infants
at 2 and 4 Months of Corrected Age (a)

                    2 Months of Corrected Age
                    Infants With VLBW
                    Low GA                    High GA
                    (n=9)                     (n= 13)

Variable            X      SD     Range       X      SD     Range

Kick frequency      14.0   10.2      0-42     22.7   14.8      3-59
  (cycles/min)
Kick amplitude (o)  34.4   10.2   22.6-59.7   39.3   12.3   12.7-61.1
Phase duration (s)
  Flexion phase      0.49   0.15   0.28-0.76   0.48   0.12   0.31-0.71
  Intrakickpause     0.27   0.38   0.02-1.64   0.74   1.22   0.01-4.71
    phase
  Extension phase    0.48   0.13   0.28-0.79   0.56   0.16   0.35-1.12
  Interkick pause    2.70   4.31   0.02-16.8   1.27   0.90   0.04-3.07
    phase

                                              4 Months of Corrected Age
                    Full-Term                 Infants with VLBW
                    Infants                   Low GA
                    (N=22)                    (n=9)

Variable            X      SD     Range        X      SD     Range

Kick  frequency     21.1   14.0      3-72      28.6   15.7      3-51
  (cycles/min)
Kick amplitude (o)  41.4   10.9   14.9-76.6    53.1    9.9   37.4-72.6
Phase duration (s)
  Flexion phase      0.50   0.12   0.31-0.92    0.54   0.14   0.37-0.79
  Intrakick pause    0.74   1.86   0.01-12.07   1.82   3.92   0.02-15.3
    phase
  Extension phase    0.54   0.14   0.24-0.88    0.50   0.16   0.31-0.83
  Interkick pause    1.57   1.71   0.05-9.00    0.62   0.62   0.03-1.94
    phase

                                              Full-Term
                    High GA                   Infants
                    (n=13)                    (N=22)

Variable            X      SD     Range       X      SD     Range

Kick frequency      14.7    9.0      0-33     16.5   13.0      3-60
   (cycles/min)
Kick amplitude (o)  48.1   12.8   23.7-80.9   52.1   11.9   29.3-77.6
Phase duration (s)
  Flexion phase      0.70   0.18   0.33-1.10   0.75   0.26  0.32-1.63
  Intrakickpause     2.29   3.60   0.02-13.67  2.89   4.14  0.10-16.38
    phase
  Extension phase    0.56   0.16   0.34-1.00   0.56   0.33  0.25-2.45
  Interkickpause     1.28   1.12   0.03-3.90   1.27   1.41  0.02-5.15
    phase

(a) GA=gestational age.
Table 4.
Number and Percentage of Preterm Infants With Very Low Birth Weight
(VLBW) and Full-Term Infants Who Adopted Alternate, Unilateral, and
Synchronous Kicks as the Predominant Interlimb Coordination Pattern at
2 and 4 Months of Corrected Age

                    2 Months of Corrected Age
                    Infants With VLBW           Full-Term
Predominant         Low GA      High GA          Infants
Pattern             (n=9)       (n= 13)          (N=22)

Alternate kicks     2 (22.2%)   2 (15.4%)         4 (18.2%)
Unilateral kicks    6 (66.7%)   4 (30.8%)        10 (45.4%)
Synchronous kicks   1 (11.1%)   7 (53.8%)         8 (36.4%)

                    4 Months of Corrected Age
                    Infants With VLBW           Full-Term
Predominant         Low GA      High GA         Infants
Pattern             (n=9)       (n= 13)         (N=22)

Alternate kicks     1 (11.1%)   5 (38.5%)        2 (9.1%)
Unilateral kicks    0 (0%)      2 (15.4%)        4 (18.2%)
Synchronous kicks   8 (88.9%)   6 (46.1%)       16 (72.7%)

* GA=gestational age.

This study was approved by the Institutional Review Committee of the College of Medicine, National Taiwan University.

Part of the data reported in this paper were presented at the Twelfth Biennial International Conference on Infant Studies; July 16-20, 2000; Brighton, United Kingdom.

This work was supported by a grant from the National Health Research Institute (DOH 88-HR-619) of the Department of Health in Taiwan.

This article was submitted January 9, 2001, and was accepted July 24, 2001.

References

(1) Ornstein M, Ohlsson A, Edmonds J, et al. Neonatal follow-up of very low birthweight/extremely low birthweight infants to school age: a critical overview. Acta Pediatr Scand. 1991;80:741-748.

(2) Calame A, Fawer CL, Claeys V, et al. Neurodevelopmental outcome and school performance of very-low-birthweight infants at 8 years of age. Eur J Pediatr. 1986;145:461-466.

(3) Bartlett D, Piper MC. Neurodevelopment of preterm infants through the first year of life: implications for physical and occupational therapists. Physical & Occupational Therapy in Pediatrics. 1993;12:37-55.

(4) Shonkoff JP, Hauser-Cram P. Early intervention for disabled infants and their families: a quantitative analysis. Pediatrcs. 1987;80:650-658.

(5) Palisano RJ. Neuromotor and developmental assessment. In: Wilhelm IJ, ed. Physical Therapy Assessment in Early Infancy. Clinics in Physical Therapy. New York, NY: Churchill Livingstone Inc; 1993: 173-224.

(6) Prechtl HFR. Qualitative changes of spontaneous movements in fetus and preterm infants are a marker of neurological dysfunction. Early Hum Dev. 1990;23:151-158.

(7) Einspieler C. Abnormal spontaneous movements in infants with repeated sleep apnoeas. Early Hum Dev. 1994;36:31-48.

(8) Bos AF, Martijn A, van Asperen RM, et al. Qualitative assessment of general movements in high-risk preterm infants with chronic lung disease requiring dexamethasone therapy. J Pediatr. 1998; 132: 300-306.

(9) Geerdink JJ, Hopkins B, Beek WJ, Heriza CB. The organization of leg movements in preterm and full-term infants after term age. Dev Psychobiol. 1996;29:335-351.

(10) van der Heide JC, Paolicelli PB, Boldrini A, Cioni G. Kinematic and qualitative analysis of lower-extremity movements in preterm infants with brain damage. Phys Ther. 1999;79:546-557.

(11) Yokochi K, Inukai K, Hosoe A, et al. Leg movements in the supine position of infants with spastic diplegia. Dev Med Child Neurol. 1991;33: 903-907.

(12) Albers S, Jorch G. Prognostic significance of spontaneous motility in very immature preterm infants under intensive care treatment. Biol Neonate. 1994;66:182-187.

(13) Thelen E, Fisher DM. The organization of spontaneous leg movements in newborn infants. Journal of Motor Behavior. 1981;15:353-377.

(14) Thelen E, Ridley-Johnson R, Fisher DM. Shifting patterns of bilateral coordination and lateral dominance in the leg movements of young infants. Dev Psychobiol. 1983;16:29-46.

(15) Thelen E. Developmental origins of motor coordination: leg movements in human infants. Dev Psychobiol. 1985;18:1-22.

(16) Thelen E, Skala KD, Kelso JAS. The dynamic nature of early coordination: evidence from bilateral leg movements in young infants. Dev Psychol. 1987;23:179-186.

(17) Thelen E, Bradshaw G, Ward JA. Spontaneous kicking in month-old infants: manifestation of a human central locomotor program. Beh Neur Biol. 1981;32:45-53.

(18) Harris SR, Heriza CB. Measuring infant movement: clinical and technological assessment techniques. Phys Ther. 1987;67:1877-1880.

(19) Winter DA. Kinematics. In: Winter DA, ed. Biomechanics and Motor Control of Human Movement. 2nd ed. New York, NY: John Wiley & Sons Inc; 1990:11-50.

(20) Fetters L. Measurement and treatment in cerebral palsy: an argument for a new approach. Phys Ther. 1991;71:244-247.

(21) Heriza CB. Organization of leg movements in preterm infants. Phys Ther. 1988;68:1340-1346.

(22) Heriza CB. Comparison of leg movements in preterm infants at term with healthy full-term infants. Phys Ther. 1988;68:1687-1693.

(23) Droit S, Boldrini A, Cioni G. Rhythmical leg movements in low-risk and brain-damaged preterm infants. Early Hum Dev. 1996;44:201-213.

(24) Vaal J, van Soest AJ, Hopkins B. Spontaneous kicking behavior in infants: age-related effects of unilateral weighting. Dev Psychobiol. 2000;36:111-122.

(25) Vaal J, van Soest AJ, Hopkins B, et al. Development of spontaneous leg movements in infants with and without periventricular leukomalacia. Exp Brain Res. 2000;135:94-105.

(26) Prechtl HFR. Continuity and change in early neural development. In: Prechtl HFR, ed. Continuity of Neural Functions From Prenatal to Postnatal Life. London, England: Spastic International Medical Publications; 1984:1-15.

(27) Hsieh TT, Hsu JJ, Chen CJ, et al. Analysis of birth weight and gestational age in Taiwan. J Formos Med Assoc. 1991;90:382-387.

(28) Skidmore MD, Rivers A, Hack M. Increased risk of cerebral palsy among very low-birthweight infants with chronic lung disease. Dev Med Child Neurol. 1990;32:325-332.

(29) Papile LA, Burstein J, Burstein R, Koffier H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr. 1978;92: 529-534.

(30) Committee for the Classification of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. Arch Ophthalmol. 1984;102:1130-1134.

(31) Johnson A, Goddard O, Ashurst H. Is late walking a marker of morbidity? Arch Dis Child. 1990;65:486-488.

(32) Jeng SF, Yau KIT, Liao HF, et al. Prognostic factors for walking attainment in very low-birthweight preterm infants. Early Hum Dev. 2000;59:159-173.

(33) Brazelton TB. Neonatal Behavioral Assessment Scale. Clinics in Developmental Medicine No. 68. Philadelphia, Pa: JB Lippincott Co; 1984.

(34) Abdel-Aziz YI, Karara HM. Direct linear transformation from comparator co-ordinates into object space coordinates. In: Proceedings of the ASP/UI Symposium on Close-Range Photogrammetry. Falls Church, Va: American Society of Photogrammetry; 1971:1-18.

(35) Jensen JL, Ulrich BD, Thelen E, et al. Adaptive dynamics of the leg movement patterns of human infants, I: the effects of posture on spontaneous kicking. J Motor Beh. 1994;26:303-312.

(36) Bennett FC, Robinson NM, Sells CJ. Growth and development of infants weighing less than 800 grams at birth. Pediatrics. 1983;71: 319-323.

(37) Harris SR, Swanson MW, Andrews MS, et al. Predictive validity of the "Movement Assessment of Infants." J Dev Behav Pediatr. 1984;5: 336-342.

(38) Hopkins B, Prechtl HFR. A qualitative approach to the development of movements during early infancy. In: Prechtl HFR, ed. Continuity of Neural Functions From Prenatal to Postnatal Life. London, England: Spastic International Medical Publications; 1984:179-197.

(39) van Wulfften T, Hopkins B. Development of the infant's social competence during early face-to-face interaction: a longitudinal study. In: Prechtl HFR, ed. Continuity of Neural Functions From Prenatal to Postnatal Life. London, England: Spastic International Medical Publications; 1984:198-219.

(40) Nigg BM, Cole GK. Measuring techniques: optical methods. In: Nigg BM, Herzog W, eds. Biomechanics. New York, NY: John Wiley & Sons Inc; 1994:254-286.

(41) Touwen B. Neurological Development in Infancy. London, England: SIMP and Heinemann; 1976.

(42) Papousek H, Papousek M. Qualitative transitions in integrative processes during the first trimester of human postpartum life. In: Prechtl HFR, ed. Continuity of Neural Functions From Prenatal to Postnatal Life. London, England: Spastic International Medical Publications; 1984:220 -244.

SF Jeng, PT, ScD, is Associate Professor, School and Graduate Institute of Physical Therapy, College of Medicine, National Taiwan University, No. 7 Chun-Shan South Rd, Taipei, Taiwan ([e-mail]). Address all correspondence to Dr Jeng.

LC Chen, PT, MS, is Doctoral Student, Department of Kinesiology, University of Maryland, College Park, Md.

KIT Yau, MD, is Professor, School of Medicine, College of Medicine, Fu-Jen Catholic University, and Deputy Director of Education and Neonatologist, Cardinal Tien Hospital, Taipei, Taiwan.

All authors provided concept/research design, writing, data collection, project management, fund procurement, and facilities/equipment. Li-Chiou Chen provided data analysis, and Dr Yau provided subjects and consultation (including review of manuscript before submission). The authors thank the infants and their parents for participation in this study; Ms Hsian-Feng Chen, Ms Shiu-Ying Yu, and Ms Wen-May Tsai for their assistance in data collection; and Dr Linda Fetters, Ms Hua-Fang Liao, and Dr Tung-Wu Lu for their guidance during the study.