The Veterinary Journal 198 (2013) e70–e74
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Development of postural balance in foals Sandra Nauwelaerts ⇑, Sara R. Malone, Hilary M. Clayton McPhail Equine Performance Center, Michigan State University, East Lansing, MI, USA
a r t i c l e
i n f o
Keywords: Equine Postural sway Balance control Stability Center of pressure
a b s t r a c t This study used stabilographic analysis to measure and describe changes in stability during standing in foals from birth to 5 months of age. Stabilographic analysis was performed on newborn foals immediately after first suckling then daily until 1 week of age, weekly until 1 month of age and monthly until 5 months of age. Ground reaction force (GRF) data were collected for periods of 8 s with the foal standing on one or two force plates recording at 1000 Hz. Stabilographic variables describing the amplitude, velocity and frequency of center of pressure (COP) movements were derived from the GRF data. Amplitudes, which were initially larger in the craniocaudal direction, decreased over time in both directions, with craniocaudal amplitude becoming smaller than mediolateral amplitude by 1–2 months of age. At birth, COP velocity was larger in the craniocaudal direction, but decreased rapidly to become smaller than mediolateral velocity by 3 months of age. Mean frequency at birth was higher craniocaudally, but became similar in both directions at 2 months of age. The rapid reductions in craniocaudal amplitude and velocity were thought to reflect improvements in strength and coordination of the flexor/extensor musculature. Newborn foals splay their limbs to compensate for poor control of the abductor/adductor musculature and, after the limbs assumed a vertical posture, mediolateral sway velocity increased. Ó 2013 Elsevier Ltd. All rights reserved.
Introduction Horses are precocial, which means that their young are relatively mobile at birth. Foals stand up within a few hours after birth to suckle. Neonates of precocial species have been described as having righting reflexes and displaying adult-like postural abilities and coordination (Fox, 1964; Lelard et al., 2006). Research on development of motor control has mainly focused on altricial species. The assumption of adult-like features might explain why the development of postural ability in precocial species has been understudied, but this assumption has never been explicitly tested. Postural balance is defined as the ability to stand in a comfortable posture with minimal movements of the body segments. In order to maintain static balance, the body’s center of mass (COM) must be maintained within the limits of the base of support, which are determined by the contacts with the ground. Mechanically, balance is achieved by counteracting gravitational forces on the COM with ground reaction forces (GRF) under the feet. The COM moves constantly and these movements are monitored by visual, vestibular and somatosensory receptors (Forssberg and Nasher, 1982). The sensory input guides the muscular response, resulting in changes in the GRF distribution beneath the feet. These GRFs can be measured using one or more synchronized force plates. Movements of the body’s center of pressure ⇑ Corresponding author. Tel.: +1 32 3 2652119. E-mail address: [email protected] (S. Nauwelaerts). 1090-0233/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2013.09.036
(COP), calculated as the position of the origin of the resultant of the vertical GRF, can be plotted in the horizontal plane as a stabilogram. Derived stabilographic variables quantify performance of the postural balance system. Development of postural control requires maturation of the nervous and musculoskeletal systems and the interaction between them (Massion, 1992). It takes almost a year before human infants are able to stand independently and many additional years are required for them to develop adult-like postural control (Chen et al., 2008; Wu et al., 2009) using a combination of open-loop and closed-loop mechanisms (Collins and DeLuca, 1993). Closed-loop control is monitored by sensory feedback, while open-loop control involves ballistic correctional movement, the effectiveness of which is not known until after the movement has occurred (Riach and Starkes, 1994). Younger children appear to rely on fast but imprecise ballistic control, with a gradual change at 4–7 years of age to a more adaptable sensory feedback correction strategy. This is associated with reductions in both sway amplitude and velocity that are independent of changes in height and weight (Riach and Starkes, 1994). In sharp contrast to the body of knowledge describing the development of balance in children, little is known about development and maturation of balance in foals, which differ in several ways from human infants. The precocious development of the musculoskeletal system allows foals to stand and move at rapid speed within a few hours after birth (Adams and Mayhew, 1984; Acworth, 2003). The rectangular base of support is long
e71
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
the doorway to the foaling stall. The foal was oriented facing the mare that observed the procedure and was restrained about 1 m from her foal (Fig. 1). During data collection, one or two handlers stood close to the foal, but without any direct contact with the foal’s body. After collecting force data, the foal was returned to the stall. On subsequent occasions, the foal was either carried or walked to the force plate for data collections. From birth until each foals was 2-months-old, data were collected using a single force plate. From 3 months of age onwards, two synchronized force plates were used to accommodate the larger base of support of the growing foals. In accordance with the normal management practices at the farm, each mare was turned out with her foal daily (9.00 am to 3.00 pm) in individual paddocks during the first week. The mare and foal were then turned out full-time in a large pasture with a group of mares and foals. The foals’ hooves were trimmed at 6 week intervals starting at 2 months of age.
craniocaudally (CC) and short mediolaterally (ML), which is the opposite of the relative dimensions of the human base of support. It seems reasonable to assume from the size and shape of the base of support that balancing strategies of a foal will differ from those of a biped. This study quantified the postural balance of newborn foals and measured changes in stabilographic variables during the early days, weeks and months of life. The experimental hypothesis was that, even in this precocial species, values of stabilographic variables representing the amplitude, velocity and frequency of COP movements will decrease with age.
Data collection Materials and methods Postural stability was measured using two Bertec FP6090 force plates (Bertec Corporation) with 600 mm 900 mm 150 mm top plates, each with 900 kg load capacity recording at 1000 Hz. Each force plate had a 16-bit digital internal amplifier, with embedded calibration information to reduce cross-talk between channels. The force data were fed through a digital AM6800 amplifier (Bertec Corporation) connected to a laptop. Forces and moments applied to the force plate were measured in three dimensions. During data recording, the foals had to stand motionless without stepping or visual movement of the head and neck. A digital camcorder was positioned on the lateral side of the force plate to enable qualitative screening of the trials. Data were recorded on the day of birth, daily for the first week, weekly for the first month and monthly until the foals were weaned at 5 months of age. Each phase of data recording consisted of at least three trials, each with a minimal duration of 10 s. The duration of the trials was standardised to 8 s. Movements of the COP of the resultant GRF were calculated and displayed in a stabilogram. The origin of the stabilogram was defined as the mean value of all data points in the craniocaudal (CC) and mediolateral (ML) directions, which corresponded with the longitudinal and transverse axes of the force plate(s). CC and ML amplitudes of the COP were determined from the stabilogram as the differences between maximal and minimal values in the two perpendicular horizontal directions. CC and ML velocities of the COP (mean and standard deviation) were calculated as the first derivative of the COP position in the CC and ML directions, respectively, through time. Fast Fourier transformation was used to determine mean power frequency. The foal’s weight was the average vertical force during the trial.
Subjects The subjects were 12 Arabian foals that had a normal gestation and parturition. After birth, the foals were allowed to stand and suckle the mare undisturbed. After suckling, each foal was carried to the force plate, which was positioned just outside
Statistical analysis The data were linearized by log transformation and the transformed data were analysed using a mixed model ANOVA. Each trial was used as a separate measurement and no averaging was performed. Relationships between sway amplitudes and velocities were examined using Pearson’s product–moment correlation. All statistical tests were based on a probability of P < 0.05.
Results Fig. 1. Foal standing on a single force plate in preparation for data collection. During recordings, the handler stood close to the foal but had no physical contact with it.
Body weights of the foals increased through the period of study (Fig. 2). Sway amplitudes were initially larger in the CC direction
Table 1 Mean stabilographic variables (amplitude, velocity and frequency in the craniocaudal and mediolateral directions) from birth to 15 months of age (n = 12) with Tukey-B groupings. Days after birth 0 1 2 3 4 5 6 7 14 21 28 56 84 112 140
Craniocaudal amplitude
Group CC amplitude
Mediolateral amplitude
Group ML amplitude
Craniocaudal velocity
Group CC velocity
Mediolateral velocity
Group ML velocity
Craniocaudal frequency
Group CC frequency
Mediolateral frequency
Group ML frequency
48 42 41 42 38 37 36 37 26 26 27 27 23 19 15
1 1/2 3 2/3 3 3 4 3 4 4 4 4 4 4 4
62 53 53 53 43 43 41 48 33 27 29 18 6 6 5
1 1/2 1/2 1/2 2/3 2/3 2/3 1/2 3/4 4 4 5 6 6 6
32 26 23 24 21 21 23 21 16 13 13 16 25 20 15
1 2 2/3 2/3 2/3 2/3 2/3 2/3 4 4 4 4 2/3 3 4
55 59 54 57 49 46 58 55 37 28 28 16 5 5 4
1 1 1 1 1 1/2 1 1 2 3 3 4 5 5 6
0.54 0.59 0.65 0.49 0.54 0.56 0.55 0.50 0.47 0.52 0.45 0.55 0.67 0.73 0.60
1 1 1 1 1 1 1 1 1 1 1 1/2 2 2 2
0.73 0.90 0.81 0.77 0.90 0.70 0.96 0.77 0.76 0.65 0.70 0.68 0.63 0.62 0.61
1 1 1 1 1 1 1 1/2 1/2 1/2 1/2 1/2 2 2 2
Days with the same Tukey B group numbers do not differ significantly from each other. Only days belonging to separate groups without overlap differ significantly.
e72
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
Fig. 2. Mean ± standard deviation of body weight of foals (n = 12) during the first 5 months of age.
but they decreased progressively over time in both directions (Table 1). The decrease in sway amplitude was more rapid in the CC direction and it became smaller than ML amplitude by 2 months of age. Results of ANOVA (Fig. 3) showed that, although CC amplitude decreased rapidly over the first month, the data were grouped into several overlapping subsets (Table 1). There were significant
decreases at 2 and 3 months of age, after which CC amplitude did not change. For ML amplitude, there were no significant differences between days of the first week, but significant decreases occurred at 2 weeks and 4 months of age (Fig. 3). The standard deviations for both CC and ML amplitudes were largest immediately after birth and decreased markedly over time. Stabilograms from one foal from birth to 5 months of age are shown in Fig. 4. At birth, CC velocity was considerably higher than ML velocity, but it decreased significantly at 2 weeks, 3 weeks, 2 months and 3 months of age, by which time it was lower than ML velocity. Mediolateral velocity was significantly higher on day 0 than on any other day. The post hoc tests grouped the values into over-lapping subsets during the remainder of the first week of age, then showed significant decreases from 2 weeks to 2 months of age, after which the values increased again in the third and fourth months of age (Fig. 3). Initially, mean frequencies were higher in the CC direction and the values showed little change over time. In the ML direction, frequencies did not change over the first 2 months of age, after which they became higher than in the first month. At 5 months of age, CC and ML frequencies were similar. Standard deviations for the frequency measurements remained high throughout the study.
Fig. 3. Mean ± standard deviation for the stabilographic variables describing amplitude (A and B), velocity (C and D) and frequency (E and F) in the craniocaudal (filled circles, right panels) and mediolateral (open circles, left panels) directions from birth to 15 months of age (n = 12). Results of the ANOVA and Tukey B post hoc tests are shown above each graph. Note that data points for CC and ML amplitude are similar in magnitude at 2 months and error bars for CC amplitude and velocity at 3, 4 and 5 months are smaller than the diameter of the data points.
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
e73
Fig. 4. Typical stabilograms of one example foal from birth to 5 months of age. The first trial recorded on each day is shown after trimming the data to 8 s duration and all stabilograms are shown with the same scaling on both axes, which is shown on the first stabilogram.
The CC and ML amplitudes and velocities were correlated with each other (amplitude: r = 0.734; velocity: r = 0.645) and, in each direction, amplitudes was correlated with velocity (CC: r = 0.760; ML: r = 0.848). Discussion This study has shown large age-related reductions in amplitude and velocity of COP movements in the CC direction, and smaller changes in the ML direction, which supports the experimental hypothesis. The patterns of change in amplitude and velocity were similar, which is not surprising, since these variables were correlated, as they are in human beings (Riach and Starkes, 1994). Even though sway amplitude and velocity decreased rapidly in the first 2 weeks after birth, significant decreases could still be detected after 3 months of age. Foals are able to stand unaided soon after birth and to move in a coordinated manner shortly thereafter. Foals usually stand within 2 h postpartum and are sufficiently well balanced to nurse soon afterwards (Adams and Mayhew, 1984; Acworth, 2003). Early ingestion of colostrum is necessary for passive transfer of immunoglobulins (Raidal et al., 2005), so foals in this study were not disturbed until after they had suckled. Neonatal foals are less aware of and less responsive to external stimuli than older foals, and this facilitated data collection. When restrained firmly with one arm around the chest and the other around the rump, neonatal foals become relaxed and almost cataplectic (Adams and Mayhew, 1984). This response facilitated placing foals on the force plate during the early days of life. Additionally, foals are not born with a full set of reflex responses. For example, the menace reflex develops during the first 2 weeks postpartum (Adams and Mayhew, 1984). By the time these reflexes were present, the foals were already accustomed to standing on the force plate and accepted it without resistance. Our results indicate that newborn foals were initially quite unstable, as shown by the relatively large values of COP amplitude and velocity, which are indicative of poor postural control. However, the magnitudes of these variables decreased rapidly over the first week of life and then more slowly over the following weeks and months as muscular strength and neuromotor control improved. Children show a change in strategy from open-loop control to closed-loop control at 7–8 years of age (Riach and Starkes, 1994). In closed-loop control, COP movements are monitored by sensory feedback and the inherent time delay in processing the feedback by the central nervous system results in slower COM velocity (Riach and Starkes, 1994). We suggest that neonatal foals, with their incompletely developed neuromuscular reflexes, weak musculature and inexperience in postural control, rely on open-loop control, but over the first
2 weeks of life they learn to integrate sensory input and change to a closed-loop mechanism, with consequent reductions in COP amplitude and velocity. COP amplitude, velocity and frequency were all higher in the CC direction initially, but became lower in the CC than the ML direction by 3 months of age. This may be related to the fact that equine limbs have evolved to move primarily in a parasagittal plane, which implies a large range of motion in flexion–extension. In accordance with the need to move and stabilise the joints in the parasagittal plane, the flexor/extensor musculature is well developed, although not fully functional immediately after birth. In the postnatal period, the musculature develops and hypertrophies, which affects postural stability during standing. Furthermore, the equine base of support has its longer dimension craniocaudally, which allows the COM to move farther in this direction without risk of falling. Consequently, the fact that COP amplitude and velocity were initially larger in the CC direction may reflect laxity in control in this direction, since the COM can migrate farther without leaving the base of support. Reduction of CC sway in this direction likely accompanies the development of co-contraction in flexor and extensor muscles, resulting in increased joint stiffness, which has the effect of stabilizing the body in the sagittal plane (Wang et al., 2006). A further indicator of maturation of the musculoskeletal system is that the gaits, which initially are stilted and hypermetric, become better coordinated within a few days after birth (Adams and Mayhew, 1984). We hypothesize that improved coordination skills, combined with increasing strength of the flexor–extensor musculature with age, stiffened the joints, which damped the sagittal plane movements and thus decreased CC sway amplitude and velocity. The equine base of support is narrow in the ML direction and has a relatively small abductor and adductor muscle volume to provide stability in this direction. Newborn foals splay their limbs (Adams and Mayhew, 1984; Acworth, 2003), which increases the width of the base of support and may assist in maintaining the COM within the lateral boundaries. Adoption of a splayed posture may also facilitate using the extrinsic limb musculature to adjust ML forces and sway patterns. Abduction and adduction are possible at the distal interphalangeal, hip and shoulder joints. Collateralmotion of the distal interphalangeal joints allows the hoof to remain flat on the ground with the limb abducted (Chateau et al., 2002), while the shoulder and hip joints allow limb abduction relative to the trunk. With the elbow, carpal and metacarpophalangeal joints, or the stifle, tarsal and metacarpophalangeal joints, aligned in the frontal plane, the limbs act as struts to resist transverse movements of the body. Consequently, with the limbs fixed in this orientation, ML amplitude and velocity were relatively small in neonatal foals. As in the CC direction, improvements in strength
e74
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
and coordination allow the extrinsic musculature to take over the function of providing mediolateral stability, with the limbs becoming more vertically oriented in older foals. ML velocity started to increase at 2 months of age and became significantly higher from 3 months of age onwards compared with the first month. It has been reported that a higher velocity of the COP may be indicative of a more responsive postural control system. Perhaps the smaller ML dimension of the base of support, combined with the smaller volume of abductor–adductor musculature, requires a more dynamic postural response in this direction, resulting in a switch to an open-loop ballistic control strategy, as described by Riach and Starkes (1994). Although parts of the postural control system are functional at birth in precocious species such as the horse, the more complex motor patterns used to respond to perturbations still need to undergo maturation. In children, stereotyped, automatic postural adjustments are controlled by inherent central networks, while the incorporation of sensory input is considered to require a higher level of control and a learned ability (Forssberg and Nasher, 1982). This may also be true in horses. Average sway frequency was low in newborn foals (
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Development of postural balance in foals Sandra Nauwelaerts ⇑, Sara R. Malone, Hilary M. Clayton McPhail Equine Performance Center, Michigan State University, East Lansing, MI, USA
a r t i c l e
i n f o
Keywords: Equine Postural sway Balance control Stability Center of pressure
a b s t r a c t This study used stabilographic analysis to measure and describe changes in stability during standing in foals from birth to 5 months of age. Stabilographic analysis was performed on newborn foals immediately after first suckling then daily until 1 week of age, weekly until 1 month of age and monthly until 5 months of age. Ground reaction force (GRF) data were collected for periods of 8 s with the foal standing on one or two force plates recording at 1000 Hz. Stabilographic variables describing the amplitude, velocity and frequency of center of pressure (COP) movements were derived from the GRF data. Amplitudes, which were initially larger in the craniocaudal direction, decreased over time in both directions, with craniocaudal amplitude becoming smaller than mediolateral amplitude by 1–2 months of age. At birth, COP velocity was larger in the craniocaudal direction, but decreased rapidly to become smaller than mediolateral velocity by 3 months of age. Mean frequency at birth was higher craniocaudally, but became similar in both directions at 2 months of age. The rapid reductions in craniocaudal amplitude and velocity were thought to reflect improvements in strength and coordination of the flexor/extensor musculature. Newborn foals splay their limbs to compensate for poor control of the abductor/adductor musculature and, after the limbs assumed a vertical posture, mediolateral sway velocity increased. Ó 2013 Elsevier Ltd. All rights reserved.
Introduction Horses are precocial, which means that their young are relatively mobile at birth. Foals stand up within a few hours after birth to suckle. Neonates of precocial species have been described as having righting reflexes and displaying adult-like postural abilities and coordination (Fox, 1964; Lelard et al., 2006). Research on development of motor control has mainly focused on altricial species. The assumption of adult-like features might explain why the development of postural ability in precocial species has been understudied, but this assumption has never been explicitly tested. Postural balance is defined as the ability to stand in a comfortable posture with minimal movements of the body segments. In order to maintain static balance, the body’s center of mass (COM) must be maintained within the limits of the base of support, which are determined by the contacts with the ground. Mechanically, balance is achieved by counteracting gravitational forces on the COM with ground reaction forces (GRF) under the feet. The COM moves constantly and these movements are monitored by visual, vestibular and somatosensory receptors (Forssberg and Nasher, 1982). The sensory input guides the muscular response, resulting in changes in the GRF distribution beneath the feet. These GRFs can be measured using one or more synchronized force plates. Movements of the body’s center of pressure ⇑ Corresponding author. Tel.: +1 32 3 2652119. E-mail address: [email protected] (S. Nauwelaerts). 1090-0233/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2013.09.036
(COP), calculated as the position of the origin of the resultant of the vertical GRF, can be plotted in the horizontal plane as a stabilogram. Derived stabilographic variables quantify performance of the postural balance system. Development of postural control requires maturation of the nervous and musculoskeletal systems and the interaction between them (Massion, 1992). It takes almost a year before human infants are able to stand independently and many additional years are required for them to develop adult-like postural control (Chen et al., 2008; Wu et al., 2009) using a combination of open-loop and closed-loop mechanisms (Collins and DeLuca, 1993). Closed-loop control is monitored by sensory feedback, while open-loop control involves ballistic correctional movement, the effectiveness of which is not known until after the movement has occurred (Riach and Starkes, 1994). Younger children appear to rely on fast but imprecise ballistic control, with a gradual change at 4–7 years of age to a more adaptable sensory feedback correction strategy. This is associated with reductions in both sway amplitude and velocity that are independent of changes in height and weight (Riach and Starkes, 1994). In sharp contrast to the body of knowledge describing the development of balance in children, little is known about development and maturation of balance in foals, which differ in several ways from human infants. The precocious development of the musculoskeletal system allows foals to stand and move at rapid speed within a few hours after birth (Adams and Mayhew, 1984; Acworth, 2003). The rectangular base of support is long
e71
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
the doorway to the foaling stall. The foal was oriented facing the mare that observed the procedure and was restrained about 1 m from her foal (Fig. 1). During data collection, one or two handlers stood close to the foal, but without any direct contact with the foal’s body. After collecting force data, the foal was returned to the stall. On subsequent occasions, the foal was either carried or walked to the force plate for data collections. From birth until each foals was 2-months-old, data were collected using a single force plate. From 3 months of age onwards, two synchronized force plates were used to accommodate the larger base of support of the growing foals. In accordance with the normal management practices at the farm, each mare was turned out with her foal daily (9.00 am to 3.00 pm) in individual paddocks during the first week. The mare and foal were then turned out full-time in a large pasture with a group of mares and foals. The foals’ hooves were trimmed at 6 week intervals starting at 2 months of age.
craniocaudally (CC) and short mediolaterally (ML), which is the opposite of the relative dimensions of the human base of support. It seems reasonable to assume from the size and shape of the base of support that balancing strategies of a foal will differ from those of a biped. This study quantified the postural balance of newborn foals and measured changes in stabilographic variables during the early days, weeks and months of life. The experimental hypothesis was that, even in this precocial species, values of stabilographic variables representing the amplitude, velocity and frequency of COP movements will decrease with age.
Data collection Materials and methods Postural stability was measured using two Bertec FP6090 force plates (Bertec Corporation) with 600 mm 900 mm 150 mm top plates, each with 900 kg load capacity recording at 1000 Hz. Each force plate had a 16-bit digital internal amplifier, with embedded calibration information to reduce cross-talk between channels. The force data were fed through a digital AM6800 amplifier (Bertec Corporation) connected to a laptop. Forces and moments applied to the force plate were measured in three dimensions. During data recording, the foals had to stand motionless without stepping or visual movement of the head and neck. A digital camcorder was positioned on the lateral side of the force plate to enable qualitative screening of the trials. Data were recorded on the day of birth, daily for the first week, weekly for the first month and monthly until the foals were weaned at 5 months of age. Each phase of data recording consisted of at least three trials, each with a minimal duration of 10 s. The duration of the trials was standardised to 8 s. Movements of the COP of the resultant GRF were calculated and displayed in a stabilogram. The origin of the stabilogram was defined as the mean value of all data points in the craniocaudal (CC) and mediolateral (ML) directions, which corresponded with the longitudinal and transverse axes of the force plate(s). CC and ML amplitudes of the COP were determined from the stabilogram as the differences between maximal and minimal values in the two perpendicular horizontal directions. CC and ML velocities of the COP (mean and standard deviation) were calculated as the first derivative of the COP position in the CC and ML directions, respectively, through time. Fast Fourier transformation was used to determine mean power frequency. The foal’s weight was the average vertical force during the trial.
Subjects The subjects were 12 Arabian foals that had a normal gestation and parturition. After birth, the foals were allowed to stand and suckle the mare undisturbed. After suckling, each foal was carried to the force plate, which was positioned just outside
Statistical analysis The data were linearized by log transformation and the transformed data were analysed using a mixed model ANOVA. Each trial was used as a separate measurement and no averaging was performed. Relationships between sway amplitudes and velocities were examined using Pearson’s product–moment correlation. All statistical tests were based on a probability of P < 0.05.
Results Fig. 1. Foal standing on a single force plate in preparation for data collection. During recordings, the handler stood close to the foal but had no physical contact with it.
Body weights of the foals increased through the period of study (Fig. 2). Sway amplitudes were initially larger in the CC direction
Table 1 Mean stabilographic variables (amplitude, velocity and frequency in the craniocaudal and mediolateral directions) from birth to 15 months of age (n = 12) with Tukey-B groupings. Days after birth 0 1 2 3 4 5 6 7 14 21 28 56 84 112 140
Craniocaudal amplitude
Group CC amplitude
Mediolateral amplitude
Group ML amplitude
Craniocaudal velocity
Group CC velocity
Mediolateral velocity
Group ML velocity
Craniocaudal frequency
Group CC frequency
Mediolateral frequency
Group ML frequency
48 42 41 42 38 37 36 37 26 26 27 27 23 19 15
1 1/2 3 2/3 3 3 4 3 4 4 4 4 4 4 4
62 53 53 53 43 43 41 48 33 27 29 18 6 6 5
1 1/2 1/2 1/2 2/3 2/3 2/3 1/2 3/4 4 4 5 6 6 6
32 26 23 24 21 21 23 21 16 13 13 16 25 20 15
1 2 2/3 2/3 2/3 2/3 2/3 2/3 4 4 4 4 2/3 3 4
55 59 54 57 49 46 58 55 37 28 28 16 5 5 4
1 1 1 1 1 1/2 1 1 2 3 3 4 5 5 6
0.54 0.59 0.65 0.49 0.54 0.56 0.55 0.50 0.47 0.52 0.45 0.55 0.67 0.73 0.60
1 1 1 1 1 1 1 1 1 1 1 1/2 2 2 2
0.73 0.90 0.81 0.77 0.90 0.70 0.96 0.77 0.76 0.65 0.70 0.68 0.63 0.62 0.61
1 1 1 1 1 1 1 1/2 1/2 1/2 1/2 1/2 2 2 2
Days with the same Tukey B group numbers do not differ significantly from each other. Only days belonging to separate groups without overlap differ significantly.
e72
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
Fig. 2. Mean ± standard deviation of body weight of foals (n = 12) during the first 5 months of age.
but they decreased progressively over time in both directions (Table 1). The decrease in sway amplitude was more rapid in the CC direction and it became smaller than ML amplitude by 2 months of age. Results of ANOVA (Fig. 3) showed that, although CC amplitude decreased rapidly over the first month, the data were grouped into several overlapping subsets (Table 1). There were significant
decreases at 2 and 3 months of age, after which CC amplitude did not change. For ML amplitude, there were no significant differences between days of the first week, but significant decreases occurred at 2 weeks and 4 months of age (Fig. 3). The standard deviations for both CC and ML amplitudes were largest immediately after birth and decreased markedly over time. Stabilograms from one foal from birth to 5 months of age are shown in Fig. 4. At birth, CC velocity was considerably higher than ML velocity, but it decreased significantly at 2 weeks, 3 weeks, 2 months and 3 months of age, by which time it was lower than ML velocity. Mediolateral velocity was significantly higher on day 0 than on any other day. The post hoc tests grouped the values into over-lapping subsets during the remainder of the first week of age, then showed significant decreases from 2 weeks to 2 months of age, after which the values increased again in the third and fourth months of age (Fig. 3). Initially, mean frequencies were higher in the CC direction and the values showed little change over time. In the ML direction, frequencies did not change over the first 2 months of age, after which they became higher than in the first month. At 5 months of age, CC and ML frequencies were similar. Standard deviations for the frequency measurements remained high throughout the study.
Fig. 3. Mean ± standard deviation for the stabilographic variables describing amplitude (A and B), velocity (C and D) and frequency (E and F) in the craniocaudal (filled circles, right panels) and mediolateral (open circles, left panels) directions from birth to 15 months of age (n = 12). Results of the ANOVA and Tukey B post hoc tests are shown above each graph. Note that data points for CC and ML amplitude are similar in magnitude at 2 months and error bars for CC amplitude and velocity at 3, 4 and 5 months are smaller than the diameter of the data points.
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
e73
Fig. 4. Typical stabilograms of one example foal from birth to 5 months of age. The first trial recorded on each day is shown after trimming the data to 8 s duration and all stabilograms are shown with the same scaling on both axes, which is shown on the first stabilogram.
The CC and ML amplitudes and velocities were correlated with each other (amplitude: r = 0.734; velocity: r = 0.645) and, in each direction, amplitudes was correlated with velocity (CC: r = 0.760; ML: r = 0.848). Discussion This study has shown large age-related reductions in amplitude and velocity of COP movements in the CC direction, and smaller changes in the ML direction, which supports the experimental hypothesis. The patterns of change in amplitude and velocity were similar, which is not surprising, since these variables were correlated, as they are in human beings (Riach and Starkes, 1994). Even though sway amplitude and velocity decreased rapidly in the first 2 weeks after birth, significant decreases could still be detected after 3 months of age. Foals are able to stand unaided soon after birth and to move in a coordinated manner shortly thereafter. Foals usually stand within 2 h postpartum and are sufficiently well balanced to nurse soon afterwards (Adams and Mayhew, 1984; Acworth, 2003). Early ingestion of colostrum is necessary for passive transfer of immunoglobulins (Raidal et al., 2005), so foals in this study were not disturbed until after they had suckled. Neonatal foals are less aware of and less responsive to external stimuli than older foals, and this facilitated data collection. When restrained firmly with one arm around the chest and the other around the rump, neonatal foals become relaxed and almost cataplectic (Adams and Mayhew, 1984). This response facilitated placing foals on the force plate during the early days of life. Additionally, foals are not born with a full set of reflex responses. For example, the menace reflex develops during the first 2 weeks postpartum (Adams and Mayhew, 1984). By the time these reflexes were present, the foals were already accustomed to standing on the force plate and accepted it without resistance. Our results indicate that newborn foals were initially quite unstable, as shown by the relatively large values of COP amplitude and velocity, which are indicative of poor postural control. However, the magnitudes of these variables decreased rapidly over the first week of life and then more slowly over the following weeks and months as muscular strength and neuromotor control improved. Children show a change in strategy from open-loop control to closed-loop control at 7–8 years of age (Riach and Starkes, 1994). In closed-loop control, COP movements are monitored by sensory feedback and the inherent time delay in processing the feedback by the central nervous system results in slower COM velocity (Riach and Starkes, 1994). We suggest that neonatal foals, with their incompletely developed neuromuscular reflexes, weak musculature and inexperience in postural control, rely on open-loop control, but over the first
2 weeks of life they learn to integrate sensory input and change to a closed-loop mechanism, with consequent reductions in COP amplitude and velocity. COP amplitude, velocity and frequency were all higher in the CC direction initially, but became lower in the CC than the ML direction by 3 months of age. This may be related to the fact that equine limbs have evolved to move primarily in a parasagittal plane, which implies a large range of motion in flexion–extension. In accordance with the need to move and stabilise the joints in the parasagittal plane, the flexor/extensor musculature is well developed, although not fully functional immediately after birth. In the postnatal period, the musculature develops and hypertrophies, which affects postural stability during standing. Furthermore, the equine base of support has its longer dimension craniocaudally, which allows the COM to move farther in this direction without risk of falling. Consequently, the fact that COP amplitude and velocity were initially larger in the CC direction may reflect laxity in control in this direction, since the COM can migrate farther without leaving the base of support. Reduction of CC sway in this direction likely accompanies the development of co-contraction in flexor and extensor muscles, resulting in increased joint stiffness, which has the effect of stabilizing the body in the sagittal plane (Wang et al., 2006). A further indicator of maturation of the musculoskeletal system is that the gaits, which initially are stilted and hypermetric, become better coordinated within a few days after birth (Adams and Mayhew, 1984). We hypothesize that improved coordination skills, combined with increasing strength of the flexor–extensor musculature with age, stiffened the joints, which damped the sagittal plane movements and thus decreased CC sway amplitude and velocity. The equine base of support is narrow in the ML direction and has a relatively small abductor and adductor muscle volume to provide stability in this direction. Newborn foals splay their limbs (Adams and Mayhew, 1984; Acworth, 2003), which increases the width of the base of support and may assist in maintaining the COM within the lateral boundaries. Adoption of a splayed posture may also facilitate using the extrinsic limb musculature to adjust ML forces and sway patterns. Abduction and adduction are possible at the distal interphalangeal, hip and shoulder joints. Collateralmotion of the distal interphalangeal joints allows the hoof to remain flat on the ground with the limb abducted (Chateau et al., 2002), while the shoulder and hip joints allow limb abduction relative to the trunk. With the elbow, carpal and metacarpophalangeal joints, or the stifle, tarsal and metacarpophalangeal joints, aligned in the frontal plane, the limbs act as struts to resist transverse movements of the body. Consequently, with the limbs fixed in this orientation, ML amplitude and velocity were relatively small in neonatal foals. As in the CC direction, improvements in strength
e74
S. Nauwelaerts et al. / The Veterinary Journal 198 (2013) e70–e74
and coordination allow the extrinsic musculature to take over the function of providing mediolateral stability, with the limbs becoming more vertically oriented in older foals. ML velocity started to increase at 2 months of age and became significantly higher from 3 months of age onwards compared with the first month. It has been reported that a higher velocity of the COP may be indicative of a more responsive postural control system. Perhaps the smaller ML dimension of the base of support, combined with the smaller volume of abductor–adductor musculature, requires a more dynamic postural response in this direction, resulting in a switch to an open-loop ballistic control strategy, as described by Riach and Starkes (1994). Although parts of the postural control system are functional at birth in precocious species such as the horse, the more complex motor patterns used to respond to perturbations still need to undergo maturation. In children, stereotyped, automatic postural adjustments are controlled by inherent central networks, while the incorporation of sensory input is considered to require a higher level of control and a learned ability (Forssberg and Nasher, 1982). This may also be true in horses. Average sway frequency was low in newborn foals (
