Cameron, Emma C. PhD, PT; Maehle, Valerie PhD, PT
Follow-up studies from early infancy to school age have found that the infant born very preterm and very low birth weight (VPT-VLBW) is at “high risk” for developmental morbidities.1–3 These morbidities range from cerebral palsy to minor clumsiness and coordination deficits.1–3 As a result of these outcomes, early developmental intervention programs have been designed to commence as early as medically feasible.4–6 The aim of these programs is to optimize the motor development of the infant born VPT-VLBW.4–7
The motor components of these developmental interventions include the use of postural support and handling procedures to encourage flexor development, muscle balance, and appropriate developmental behaviors for the infant’s gestational age.4–7 These components are viewed as important because investigators studying infants born preterm (PT) and infants born full term (FT) and considered healthy have reported differences in muscle power, tone and movement at term age.8–12 These investigators have reported that infants born PT exhibit weaknesses in items measuring active or reflex responses of the flexor muscles, have more extended postures, and less head control compared with infants born FT and healthy.
However, the results from these studies are limited in their applicability today because only one10 was conducted within the last decade. Numerous changes in nursing and medical care have occurred in recent years, which may have affected the neurobehavioral development of infants born PT in comparison with their peers born FT. Therefore, updated research comparing the motor performance of the infant born VPT-VLBW is required. In addition, the investigators that compared infants born PT and FT at term age8–12 used assessment methods whereby the examiner observed or elicited a motor response and documented the infant’s response in gross categories using a Likert scale. The development of a more objective method of measuring infant’s postures and movements would provide precise information regarding motor behaviors and aid in the development and evaluation of more-specific intervention programs for infants born VPT-VLBW.
The first aim of this study was to compare active motor items: head lag on pull to sit, head righting on pull to sit, head control in sitting, and hip flexion in prone between a group of infants born VPT-VLBW with normal neurological examinations and a group of infants born FT and healthy. The second aim of the study was to evaluate a more objective method of assessing motor items: head lag on pull to sit, head righting on pull to sit, head control in sitting and hip flexion in prone measured using video recordings and computer software.
The active motor items were selected after reviewing the motor items included in commonly used neonatal assessments. These motor items are included in a number of validated developmental examinations used with infants born PT and FT at term age.13–17 The criterion validity of these motor items has been established based on their predictive value;18 they also have demonstrated a longitudinal developmental progression indicating that they are measures of motor maturity.19,20 These four motor items could be measured using the equipment available and were thought to best measure the infant’s flexor tone and muscle balance. Measuring these items would therefore establish if the differences in flexor tone and muscle balance established previously by other researchers comparing infants born PT and FT8–12 are still evident today and when using a more objective measurement method.
A sample of 34 infants born VPT-VLBW was recruited within three weeks of their admission to the neonatal unit at Aberdeen Maternity Hospital. The recruitment period of the study was from April 1, 1998, to March 31, 2001. During the study period, a randomized clinical trial was being conducted in the neonatal unit to investigate the effectiveness of physiotherapy developmental interventions. None of the infants recruited to this study had received any physiotherapy or any method of developmental care.
The selection criteria were as follows: less than 1500 g less than 32 weeks’ gestational age, greater than 24 weeks’ gestational age, inborn and resident in Grampian, classified as neurologically normal at 18 months corrected age (CA), absence of congenital or musculoskeletal abnormalities, absence of drug withdrawal symptoms at birth and family unknown to the social work department. Four of the 34 infants recruited at birth did not have a normal neurological examination at 18 months’ CA and so were excluded from the final analysis. In addition, one infant was discharged unexpectedly early and so had no term age assessment available (Table 1).
Twenty-two singletons who were healthy and born FT were recruited within 48 hours of birth (Table 1). They had been born by spontaneous vaginal delivery, had a birth weight of greater than 2500 g, an Apgar score of greater than eight, and a normal examination as conducted by the pediatrician on the first postnatal day. The infants were all verified by the health visitor as developing within normal limits when contacted by the researcher at eight months of age. Informed consent was obtained from all parents before participation in the study. The Grampian Ethics Committee approved the study prior to commencement.
At term age, head lag on pull to sit, head righting on pull to sit, head control in sitting, and hip flexion in prone for the infants born VPT-VLBW and infants born FT were recorded. Head righting to body axis alignment on pull to sit is included in the classification of head control on pull to sit used by Lacey et al19(p. 203) and is observed as a response to align the head to the trunk through the 90-degree arc of the pull to sit maneuver. This measure was included because we thought that it may be sensitive to small changes in neck flexor muscle strength.
A Panasonic slim video camcorder, model number MX5B-NV, was used to record the motor items. The recordings were analyzed using computer software “video capture” (Application version 1.01), the MIE “digitizer,” and the MIE “analyzer” (MIE Medical Research Ltd, UK). Video capture allows an examiner to select single frames of movement to be captured and stored in the computer memory. The MIE digitizer is a motion-digitizing package designed to mark the joint angles from the single frames captured by the video capture program. The MIE analyzer is a computer package designed to calculate these joint angles exactly in degrees. This software has been developed for 2D motion analysis.
The infants born VPT-VLBW were assessed between 37 and 42 weeks postconceptional age by one of two pediatric physical therapists (tester one and tester two). The therapists who conducted the examinations were trained in the test procedure, including infant position, marker placement, infant handling, and behavior.
Two pediatric physical therapists were trained in camcorder operations, camcorder position, and infant position in the view finder to enable them to carry out the video recordings of the examination (tester two and tester three). The video camcorder was hand held for the recording of the examinations with the VPT-VLBW group as limitations of space and safety concerns in the neonatal unit did not allow for the camera to be mounted onto a tripod stand.
The FT group was assessed within four days of birth by a final year physical therapy student with the video camcorder mounted onto a tripod stand. Placement of the video camcorder for the recording of the examination was such that the infant’s body was in the center of the viewfinder and the camera lens was a distance of 1m from the infant.
The videos were obtained half way between feeds with infants in an awake, alert state (state four).21 The infant was undressed for the duration of the examination and placed on a mattress covered by soft flannelette. The examination rooms were all well lit and temperature controlled. Anatomical landmarks: the greater trochanter, the lateral femoral condyle, and a midpoint on the chest between the axilla and iliac crest were located on one side of the infant’s body and marked with color stickers 8 mm in diameter. The side of the infant’s body to be marked and recorded was selected at random.
During the examination the infant was placed in supine. From supine, with the head midline, the infant was pulled through 90 degrees into a sitting position with the examiner’s thumbs in the infant’s palms to elicit a traction response. This procedure was performed only once for each infant.
The infant was then supported in the upright sitting position, with the examiners hands placed anterior and posterior around the trunk. The index finger of the right hand was held upright to support the infants head posteriorly. Each infant was supported in sitting until three attempts had been made to sustain head control in this position. An “attempt” was defined as when the infant actively held the head on the body without resting posteriorly on the examiners finger or anteriorly on the chest. If the infant made no attempt to actively bring the head to the upright position the examiner tilted the infant anteriorly to facilitate the head to the upright. If the infant did not move the head forward in response to this, or the head immediately fell forward to the chest this was counted as one attempt and the procedure was repeated twice more to allow for an active response.
Hip flexion was examined with the infant placed in a symmetrical prone posture with hands by the head and head turned to the preferred side. The “preferred side” was the side to which the infant turned and rested the head when placed in prone. The infant was recorded in prone until at least three hip flexor movements had been observed.
To obtain the four measurements the videotape of the examination was played frame by frame on the “video capture” program. Frames could be played forward and back repeatedly, if required, to enable tester one to select the appropriate frame for measurement using the “digitizer” program. Tester one measured all of the infants’ motor items.
The frame at which the infant’s head initially lifted off the mattress was selected by video capture for measurement. Two intersecting lines were marked using the digitizer program to provide the angle for head lag; the line of Frankfurt (Fig. 1)22 and a line parallel to the line of the trunk. The lines intersected at the external auditory meatus. The line of Frankfurt runs through the lateral aspect of the occipital bone, the upper margin of the external auditory meatus, and the anterior margin of the floor of the orbit.22 The line of Frankfurt was marked on the selected frame using these anatomical landmarks. The line of Frankfurt is an anatomical line considered as the standard horizontal plane for the orientation of the head.23 The parallel line marking the orientation of the trunk was marked along a line running through the external auditory meatus and parallel to the anatomical markers locating the mid axial point of the body and the greater trochanter. This line was judged using a ruler.
The angle measured using the analyzer program was the angle of intersection between the line of Frankfurt22 (Fig. 1) and the line parallel to the line of the trunk through the external auditory meatus. If the infant had no head lag and the head was in line with the body then this angle was 90 degrees.
Head Righting to Body Axis Alignment on Pull to Sit.
The frame selected from the video capture for measurement was during the pull to sit movement, when the infant actively flexed the head forward in an attempt to bring the head into alignment with the trunk. Two intersecting lines were then marked over the infant using the digitizer program. A line running parallel to the table surface and through the marker of the greater trochanter was marked using a ruler to aid judgment. A second line marked the orientation of the trunk and was located by the greater trochanter and mid axial markers.
The head righting angle was measured by the analyzer program as the angle of intersection between a line running through the axilla, the axial point of the trunk and the greater trochanter and a line running parallel to the supporting surface (Fig. 2). The anterior angle was measured. This measurement of the degree of head control on pull to sit has been previously validated by Lacey et al.19
Head control was measured by tester one using a frame by frame count from the video recording. Counting of frames commenced when the infant’s head was unsupported by the examiner and was judged to be within 20 degrees of the vertical line of the body. The counting of frames continued while the head remained unsupported and between 20 degrees anterior and posterior of the vertical line of the body. The position of the head on the body was visually judged by tester one.
One frame was equivalent to 1/50 of a second. The number of frames were counted by the tester and converted to seconds. Out of the three attempts, the longest time the infant held his/her head unsupported and within 20 degrees of the vertical line of the body was documented.
Using video capture, the frame was selected for measurement when the hip was moving in a neutral position into flexion and the pelvis was at the maximal point of elevation. Once the pelvis started to depress, the hip was considered to be abducted and externally rotated and was not considered to be in a position of “true” hip flexion. The frame of maximal hip flexion was selected from each of the three hip flexor movements recorded.
The angle of hip flexion was marked using the “digitizer” program from the trunk marker, through the greater trochanter marker to the lateral femoral condyle marker (Fig. 3). The smallest angle of hip flexion calculated by the MIE analyzer from the three frames was documented for each infant.
Twenty-nine head control and hip flexion measures were available for analysis from the group of infants born VPT-VLBW, with 26 head righting measures and 28 head lag measures available. Missing data for the head righting and head lag measures were caused by video error. Measures for the four motor items were obtained for all 22 infants born FT. Head righting on pull to sit was compared between the two groups using Student t test because the data fulfilled the criteria for parametric testing. The other measures were compared using the Mann-Whitney test because of the skewed nature of the data. Data distributions will be described in the results. The intrarater reliability of the digitizing method was measured using the intraclass correlation coefficient, ICC (3,1) All tests were two-tailed and a p value of <0.05 was regarded as significant. All analyses were performed using the Statistical Package for Social Sciences (SPSS) version 10.0 (SPSS Institute, Chicago, Ill).
Reliability of the Method of Assessment
Tester one conducted the measurement taking for all infants. Tester one’s intrarater reliability was assessed using a random selection of 10 PT infants and five FT infants. The intra-rater reliability (ICC [3,1]) for the four motor items was as follows; head lag, (ICC = 0.77), head righting (ICC = 0.99), hip flexion (ICC = 0.997), head control (ICC = 0.97).
Differences Between the PT and FT Infant at Term
Comparison of the four items between the infants born VPT-VLBW and infants born FT, at term, revealed that the infants born VPT-VLBW had less head lag (median 40.8 degrees, Interquartile range [IQR] 34.5 to 60.9 degrees) than the infants born full-term (median 26.6 degrees, IQR 21.4 to 54.9 degrees; p < 0.005; Table 2).
The infants born FT had more mature head righting responses (median 129.2 degrees, IQR 117.7 to 149.6 degrees) greater head control (median 0.9 seconds, IQR 0.2 seconds to 2.6 seconds), and greater hip flexion (median 80.5 degrees, IQR 67.6 degrees to 98 degrees) compared with the infants born VPT-VLBW; 119.8 degrees (IQR 93.8 degrees to 141.9 degrees), 0.5 seconds (IQR 0.3 seconds to 1.5 seconds), and median 89°(IQR 71.7 degrees to 114.8 degrees), respectively (Table 2). However, the differences between the two groups for these motor items were small and did not reach significance (Table 2).
There was a large variability within both groups for all four motor items examined, which is illustrated by both the IQRs and the ranges provided in Table 2. The data for both groups of infants for head lag, head control, and hip flexion were positively skewed, illustrating that the majority of the infants had poor head lag and head control but a good range of hip flexion in prone. The data for head righting were approximately normally distributed for both groups of infants.
Using video and digitizing computer software, we found that head righting on pull to sit, head control in sitting, and hip flexion in prone were similar for a group of infants born VPT-VLBW and developing normally and a group of infants born FT and healthy. Only head lag on pull to sit differed between the groups, with the infants born VPT-VLBW having less head lag on pull to sit. The level of reliability achieved with this method of examination supports the use of video and computer software in accurately measuring head control, head lag, head righting and hip flexion, at term age. This method of assessment requires minimal training and is safe to use with infants born VPT-VLBW who still require special care.
Active Motor Items of Infants Born PT, VLBW, and FT
The results of the current study are dissimilar to the majority of reports of previous research in this field in which the infant born FT demonstrated more mature responses on developmental tests at term compared to the infant born PT,8–12,24,25 particularly on examination of components of flexor tonus.8–12 The small sample size increases the risk of a type II error and so this result requires verification with a larger sample size. The more objective method of examination and the smaller number of motor items examined in the current study also may have had an influence on the different result obtained. Tests such as the Dubowitz Neurological Assessment, The Neonatal Behavioural Assessment Scale, and the Neonatal Neurodevelopmental Examination were used by the previous researchers in this field and provide a less objective method of measurement of motor items compared to that obtained in the current study.13,14,17 However, these tests also examine more motor items relating to different aspects of motor performance, ie, reflexes, spontaneous movement, muscle tone, and posture. More objective measures of infants born PT and FT for these other aspects of motor development may yield differences not apparent for the four motor items examined.
In addition, the fact that the infants born PT did not have weaker motor responses in comparison to the infant born FT may reflect improved outcomes for infants born PT as a result of improved obstetric and neonatal care. The majority of studies in this field have been conducted more than a decade ago.8–11 A more recent study in this field12 also found that infant’s born PT and FT demonstrate a wide range of behaviors and responses for items on the Dubowitz neonatal neurologic examination, at term age. Similar median scores were obtained for the two groups of infants on 24 of 34 items tested. The results from this recent report and the current study illustrate that in the modern era of neonatal care, differences between infants born PT and FT when examined at term age may be less pronounced than previously reported. This result has implications for physical therapists working in this field. It indicates that it may not be necessary to provide direct developmental physical therapy interventions during the neonatal period for infants born VPT-VLBW who are at “low risk” for neurological deficits. As a result direct developmental physical therapy interventions during this time period may then be directed more specifically to “high-risk” groups of infants, a practice that is current in some regions.
In the current study the infants born VPT-VLBW had less head lag compared with the infants born FT, where previous observational studies have shown the opposite to be the case.9,11,25 However, the wide range of head lag displayed by the two groups of infants reduces any clinical significance of this result. A wide range of values also was obtained for head righting, head control and hip flexion for both groups of infants demonstrating that infants born PT and FT, who are neurologically normal, display a similarly wide range of “normal” motor behaviors at term age.
The frequency distributions of the four motor items examined indicate that the majority of infants had poor head lag and head control, moderate head righting responses, and a good range of antigravity hip flexion in prone. This pattern would potentially support a caudocephalic distribution of tone for the two groups of infants, at term age. This finding is in contrast to the cephalocaudad direction of muscle tone development documented for the infant born FT26 but in agreement with research now supporting a caudocephalic direction of muscle tone development for the infant born PT.16,19,20 Further study would be required using this method of measurement to test additional motor items over repeated examinations before any conclusions regarding the maturational sequence of muscle tone development could be drawn.
Development of the Method of Assessment
This is the first research of which the author is aware to report on and compare motor items, measured using video and kinematic analysis, for a group of infants born VPT-VLBW and a group of infants born FT. However, despite the principal researcher being the only person to measure the motor items using the video capture, analyzer, and digitizer system, the accuracy of the measures were potentially affected by the number of examiners (n = 3) and operators of the video camcorder (n = 3). Consistency was not maintained because of logistical reasons. The fact that the camcorder was hand held to collect the images of the infants born PT also would potentially affect the quality of the recordings taken. These issues should be addressed in future research using this method of measurement.
In the current study we only examined, measured, and compared four motor items; further measurement of additional motor items using this method would provide further information on differences between infants born PT and FT to inform physiotherapy interventions and optimize the motor development of the infant born VPT-VLBW. Further development of this measurement tool should involve comparison with currently used validated neonatal tests to establish the tool’s validity and clinical applicability. Longitudinal follow-up research would be important to ascertain the predictive validity of this method of examination compared to more traditional neonatal examinations. These areas of future research would then establish if this more objective method of measurement could add prognostic value to current neonatal examinations.
We report on a similarly wide range of values for motor items; head lag, head righting, head control, and hip flexion for a group of infants born VPT-VLBW and a group of infants born FT who are developing typically. Indeed head lag on pull to sit was less for the infant’s born PT in comparison with the infants born FT; however, the range of values for this motor item limits any significance of this result. Using an objective method of measurement in the modern neonatal era, our findings do not support the motor differences between infants born PT and FT which have been reported by previous authors. Further development of this method of measurement is recommended to establish the validity of the method.
The authors would like to acknowledge the contribution of Gillian Copeland, who as part fulfillment of her Bsc (Hons.) Degree in physiotherapy, assisted in the data collection of the project.
1. Foulder LA, Cooke RWI. Motor, cognitive, and behavioural disorders in children born very preterm. Dev Med Child Neurol. 2003;45:97–103.
2. Sullivan MC, McGrath MM. Perinatal morbidity, mild motor delay and later school outcomes. Dev Med Child Neurol. 2003;45:104–112.
3. Colvin M, McGuire W, Fowlie PW. Neurodevelopmental outcomes after preterm birth. BMJ. 2004;329:1390–1393.
4. Cameron EC, Maehle V, Reid J. The effects of an early physical therapy intervention for very preterm, very low birth weight infants: a randomised controlled clinical trial. Pediatr Phys Ther. 2005;17.2:107–128.
5. D’Angelo LK, Unanue RA. The special care nursery. In: Campbell SK, Vanderlinden DW, Palisano RJ, eds. Physical Therapy for Children. Philadelphia, Pa: WB Saunders; 2000:840–880.
6. Girolami GL, Campbell SK. Efficacy of a neuodevelopmental treatment program to improve motor control in infants born prematurely. Pediatr Phys Ther. 1994;6:175–184.
7. Sweeney JK, Swanson MW. Low birth weight infants: neonatal care and follow-up. In: Umphred DA, eds. Neurological Rehabilitation. St. Louis, Mo: Mosby; 2001:203–259.
8. Howard J, Parmelee MD, Kopp CB, Littman B. A neurologic comparison of pre-term and full-term infants at term conceptional age. J Pediatr. 1976;88:995–1002.
9. Kurtzberg D, Herbert G, Vaughan J, et al. Neurobehavioral performance of low birthweight infants at 40 weeks conceptional age: comparison with normal full-term infants. Dev Med Child Neurol. 1979;2:590–607.
10. Palmer PG, Dubowitz LMS, Verghote M, et al. Neurological and neurobehavioural differences between preterm infants at term and full term newborn infants. Neuropediatrics 1982;13:183–189.
11. Forslund M, Bjere I. Neurological assessment of preterm infants at term conceptional age in comparison with normal full-term infants. Early Hum Dev 1983;8:195–208.
12. Mercuri E, Guzzetta A, Laroche S, et al. Neurologic examination of preterm infants at term age: comparison with term infants. J Pediatr. 2003;142:647–655.
13. Brazelton TB. Neonatal behavioural assessment scale. Clinics in Developmental Medicine. No 88. London: Blackwell Scientific Publications Ltd; 1984.
14. Allen MC, Capute AJ. Neonatal neurodevelopmental examination as a predictor of neuromotor outcome in premature infants. Pediatrics 1989;83:498–506.
15. Carmichael K, Burns Y, Gray P, et al. Neuromotor behavioural assessment of preterm infants at risk for impaired development. Aust J Physiother. 1997;43:101–107.
16. Lacey JL, Henderson-Smart DJ. Assessment of preterm infants in the intensive-care unit to predict cerebral palsy and motor outcome at six years. Dev Med Child Neurol 1998;40:310–318.
17. Dubowitz L, Dubowitz V, Mercuri E. The neurological assessment of the preterm and full term newborn infant. Clinics in Developmental Medicine. No 79. London: William Heinemann; 1999.
18. Molteno C, Gibsz P, Wallace P, et al. Neurological examination of the preterm and full term infant at risk for developmental disabilities using the Dubowitz neurological assessment. Early Hum Dev. 1995;41:167–176.
19. Lacey JL, Henderson-Smart DJ, Edwards DA, et al. The early development of head control in preterm infants. Early Hum Dev 1985;11:199–212.
20. Lacey JL, Henderson-Smart DJ, Edwards DA. A longitudinal study of early leg postures of preterm infants. Dev Med Child Neurol. 1990;32:151–163.
21. Prechtl HFR. The behavioural states of the newborn infant: a review. Brain Res. 1974;76:185–212.
22. Akessan EJ, Leob JA, Wilson-Pauwels L. Thompsons Core Textbook of Anatomy. Philadelphia: Lippincott Company; 1990.
23. Williams PL, Warwick R, Dyson M, et al. Grays Anatomy. Edinburgh: Churchill Livingstone, Edinburgh; 1989.
24. Jeng SF, Yau KIT, Teng RJ. Neurobehavioral development at term in very low-birthweight infants and normal term infants in Taiwan. Early Hum Dev 1998;51:235–245.
25. Upadhyay SK, Singh TB, Bhatia BD. Neurobehavioural assessment of newborns. Electromyogr Clin Neurophysiol. 2000;40:113–117.
26. Schloon H. Muscle activity and post natal behaviour in newborn infants. Neuropediatrie. 1976;7:384–415.
© 2006 Lippincott Williams & Wilkins, Inc.