It has been traditionally accepted that cerebral palsy cannot be diagnosed in early infancy - that is, under the age of 12 months. The reasoning has been that neurologic findings during this period (especially during the first six months of life) do not correlate well with the child's later motor development.1 Since primitive reflexes have not received clinical recognition as early motor neurodevelopmental markers, research in gross motor skills utilizing them has not been focused on during the early months of infancy.
Motor milestones have been traditionally highlighted in baby books and also in the curricula of medical schools and pediatric training programs. Yet when motor milestones are not achieved, students, interns, and residents are not taught to analyze the reason why the particular milestone has not been reached.
THE EARLIEST INDICATIONS OF SIGNIFICANT MOTOR DISABILITY
The earliest indication of the presence of a significant motor disability may be the delay in the disappearance of a primitive reflex or the presence of such a reflex to an abnormal degree.
Primitive reflexes are essentially brain-sremautomated responses that develop during gestational life and are present at birth. The majority disappear between three and six months of age. For example, the positive support reflex (PSR) appears at the eighth or ninth month of fetal life and disappears within three months after birth.2 If the reflex is present to such a degree that head movements control both upper- and lowerextremity positioning, then it is said to be present to an "obligatory" degree and is always pathologic.
Thus, the presence of an "obligatory" primitive reflex clinically signifies the presence of a severe motor disability, making the child with such a reflex highly suspect of having cerebral palsy. This clinical clue, of itself, makes it essential that the pediatrician become familiar with a few of the primitive reflexes, so that he can monitor more closely those infants in the high-risk category and make earlier referral for habilitation therapy.
The primitive reflexes are under brain-stem control and are elicited by stimulation of the sensory end organs in the muscles, joints, and labyrinths (otoliths and semicircular canals).
Delay in the appearance of postural reflexes. Many professionals, particularly physical therapists and occupational therapists who assist with the management of cerebral-palsied children, use the delay in the appearance of a child's righting and equilibrium postural responses as markers for motor delay. These postural reflexes do not appear until the latter half of the first year, however, so it became apparent that additional research was required to focus upon the primitive reflexes that are present at birth.
The three main reasons for the delay in the appearance of the righting and equilibrium responses are (1) the "hanging on" of primitive reflexes past the age at which they are physiologically present, (2) the presence of a primitive reflex to an abnormal degree, and (3) the presence of significant hypotonia. Thus, the primitive reflexes are the earliest clinical neurodevelopmental markers for the detection of cerebral palsy.
The primitive reflex profile. I am currently working on a clinical research study3 at the John F. Kennedy Institute in Baltimore in which developmental pediatricians are attempting to make a quantitation or grading of the primitive reflexes. One of the purposes of the study is to standardize a quantitation of a primitive-reflex profile, so that all professionals could grade primitive reflexes in a similar fashion. Another goal is to devise a primitive-reflex profile for each month of infancy, from birth through 12 months. Such a scale would serve as a predictive tool for the early detection of cerebral palsy.
The third goal of the study is to determine which constellation of quantitated primitive reflexes precludes the appearance of a particular motor function, such as rolling over, sitting with support, sitting unsupported, coming to sitting, pulling to standing, cruising, and walking. This knowledge would be of great clinical significance, since parents frequently ask professionals to predict when their children will arrive at the various motor milestones.
While the details of the grading system will not be discussed, it will suffice to say that the primitive reflexes are being graded on a five-point scale (0, 1+, 2+, 3 + , and 4 + ). The grading 0 indicates that the reflex is totally absent, and 4+ is indicative of an "obligatory" reflex. Within this spectrum of quantitation, the grading 2+ indicates the reflex is present to a degree normally seen at the time when it is physiologically present - e.g., a graded 2 + asymmetric tonic neck reflex (ATNR) is one normally seen in an infant of one, two, or three months. The graded 3 + reflex is one that is slightly more evident than that seen in normal infants, while al+ graded reflex is one that is present to a slightly lesser degree than normal.
This quantitated scale, then, is a method of grading of tone changes that produce extremity posturing of all four limbs. Further details regarding the quantitation system can be found elsewhere.4
PRIMITIVE REFLEXES OF MOST INTEREST TO THE PEDIATRICIAN
While there are many primitive reflexes, those that will be of most help clinically to the pediatrician are the Moro, tonic labyrinthine, asymmetric tonic neck, Galant, crossed-extension, stepping, upper placing, lower placing, and positive support reflexes. Each will be discussed in turn.
The Moro reflex (M), well known to all pediatricians, is of great clinical significance because it usually means that central nervous system insult has occurred if it is present to an excessive degree, or absent. Asymmetry might well mean the presence of a hemiparesis or a brachial palsy.
In the newborn, this reflex is elicited by holdingthe infant in horizontal supine suspension in the palm of the hand, and extending the neck 45 degrees. Later on it is elicited by lifting the infant in the supine position by the arms (thus raising the head a bit off the table) and then letting go. The response is an extension followed by abduction of the upper extremities with partial flexing of the elbows, wrists, and fingers, with the first and second digits assuming a "C" position. The physician who is familiar with the normal response will find this a most useful clinical clue.
This reflex fades at between three and six months, with the "startle" reaction appearing. The "startle" reaction differs in that it is a flexion or withdrawal of the upper extremities.
The tonic labyrinthine primitive reflex (supine [TLSJ or prone [TLP]) is a most important and useful clinical neurodevelopment sign of infancy, since its presence to a significant degree may well mean cerebral insult. The reflex, when present to an obligatory degree, is well known to pediatricians, neurologists, and neurosurgeons and is commonly seen following inflammation (meningitis) or trauma (cerebral accident) of the central nervous system. It is puzzling to see professionals use such terms as "opisthotonic," "decorticate," or "decerebrate" posturing to describe the phenomenon without apparently realizing that the terms all describe the tonic labyrinthine primitive reflex.
Physical and occupational therapists will work to inhibit this and other primitive reflexes by implementing their "inhibitory and facilitatory" methodology aimed at facilitating the appearance of postural reflexes and subsequent motor function.
The TLS is elicited by extending the infant's neck and head 45 degrees from the midline by placing a hand between the scapulae, thus stimulating the otoliths of the inner ear. This produces a pattern of extension of the upper extremities that is manifested by shoulder retraction, resulting in secondary flexion of the upper extremities. The lower extremities assume slight extensor posturing. There can be two types of the TLS, however. In one, there is primarily shoulder retraction with positional flexion of the upper extremities at the elbows, referred to as "decorticate" posturing; in the other response, the upper extremities assume full extension with pronation, referred to as "decerebrate" posturing. The separation of the TLS into these two types may be clinically useful, since the decerebrate posturing can mean a more ominous prognosis than the decorticate one. This observation may be helpful in prognosticating the outcome of a child with head injuries, in the event either reflex predominates initially. Research in this area is still needed.
The pattern of extension can be readily broken up by simply flexing the head, bringing the otoliths to neutral position. This is the tonic labyrinthine reflex, prone (TLP), which is readily observed in the newborn when the infant is lying in the prone position.
The TLS positioning is frequently accompanied by a "tongue thrust," with the infant making adderlike movements of the tongue. These movements significantly interfere with feeding, since the "tongue thrust" will thrust the food forward out of the mouth as well as interfere with the sucking pattern. The mother will thus have great difficulty placing the food back on the tongue, becoming somewhat despondent since the majority of the food will come out of the mouth rather than moving to the rear. By placing the child in a "flexor seat," occupational therapists can break up the extensor pattern, including the tongue thrust, and reduce the time it takes to feed the child, thus relieving the parents' frustration. The physical therapist, by reducing the flexor pattern, can improve the child's ability to move into the "derotarive" or "segmental" rolling pattern, since the infant will not be able to roll over if a significant TLS is present.
This reflex usually disappears between four and six months of age.
The asymmetric tonic neck reflex (ATNR) is a "fencing position" that has been looked upon by some psychiatrists as a behavioral characteristic indicating rejection of the mother. Ir is, however, an automated response that is normally present in all infants up to three or four months of age, usually disappearing by six months. If it is present to an excessive degree, it is indicative of central nervous system insult or dysfunction. Many clinicians pay a good deal of attention to this reflex along with the TLS.
Physical therapists can assist infants who have this reflex present to a significant degree, since when this occurs, derotative rolling is inhibited and the child is not able to roll over.
The reflex is elicited by turning the head 45 degrees to the right or left side while the infant is in the supine position; the extremities on the chin side will have less flexion, appearing to go into the position of extension while the occipital side demonstrates increased flexion. Although it is traditionally taught that there is extension of extremities in the chin side and flexion on the occipital side, it should be pointed out that the normal infant is mainly in a flexor position and thus the change in tone on the chin side is accurately designated as a "lessening of flexion" and on the occiput side as an "increase of flexion." The lower extremities will assume a similar posture, but to a lesser degree. In fact, this is usually true of all primitive responses - the tone changes are more noticeable in the upper extremities than in the lower ones.
When ATNR is present to an excessive degree, the therapist will attempt to have the child maintain the neutral (midline) position, thus equalizing the tone in all four extremities. Infants who have good intelligence may well overcome the uneven tone of posture by gradually learning that (for example) to roll over they must first look away from the side to which they wish to roll. An alert infant will also learn that it is much easier to grasp an object with the occipital hand than with the chin one, since the occipital hand - being in a more flexed position - assumes a more functional position.
This reflex is the reason why pediatricians, when examining infants, must keep the head in the midline to equalize tone in the four extremities. Otherwise, there may be mild asymmetry of the tone, deep tendon reflexes, or even asymmetry in the "plantar responses," which are physiologically extensor for the first year or two of life.
It was formerly taught that children who had the ATNR present to an excessive or "obligatory" degree would not sit unsupported or walk. However, in clinical pracrice children with this degree of the ATNR have been seen to sit unsupported and even walk. This is one reason for believing that a compilation of the graded primitive reflexes is essentia] to determine which motor functions will be precluded.
A combination of the Moro, TLS, and ATNR responses must be minimal if the child is to be able to roll over in a segmental fashion.
The presence of good cognition is another important factor that often is not fully appreciated. Intelligent infants who understand the physical limitations placed on them by the presence of these automated responses soon learn how to overcome them by manipulating their heads to either side or up and down. Those who work with motorimpaired infants soon realize that the intelligence factor has not been fully appreciated by many parents or their physicians.
The Galant reflex (G) is a "trunk incurvature" reflex that is elicited by stroking the paravertebral area from the lower thoracic to the sacral region. Trunk incurvature or concavity will occur on the side that has the stimulus applied. This reflex remains somewhat of an enigma, since when found alone to an excessive degree it does not correlate with any clinical finding. However, perhaps in conjunction with other primitive reflexes, it may well be of clinical significance. It is said to correlate with truncal instability, but this is true only when found in conjunction with other primitive reflexes.
The crossed-extension reflex (CER) or stepping reflex (SR) will be considered together, since I believe these spinal reflexes are one and the same - the stepping reflex simply being a vertical crossed-extension reflex. They occur when a stimulus is applied to one extremity, evoking a "protective-like" flexion reflex of the contralateral extremity.
The horizontal form - the crossed-extension reflex - is elicited by applying a noxious stimulus to the sole of one foot, which is held in complete extension with contralateral leg flexion (phase 1), followed by adduction (phase 2), and finally extension (phase 3), as if to push away the noxious stimulus.
The vertical form - the stepping reflex - can be elicited by holding the infant in the vertical position; when the sole of the foot is touched to the ground (hallucal stimulation), the contralateral foot flexes, adducts, and extends. While the contralateral foot is flexing and adducting, the examiner immediately turns the infant so that when contralateral extension takes place, it receives the weight, thus producing a "walking" or "stepping" response. MacKeith'' has shown that by extending the head, the walking response may be continued for a longer period. The reason for this is that byextending the head and putting the otoliths in 45 degree extension, the tonic labyrinthine response is elicited, giving rise to extension of the lower extremities and enhancing the stepping response.
Both of these reflexes are present at birth and disappear between the fourth and sixth weeks of life.
Placing responses. Both the lower-extremity and upper-extremity placing reflexes are most valuable for making a quick evaluation of upper and lower-extremity functioning, since their absence of asymmetry may be indicative of a motor disability. The lower-extremity placing reflex (LPR) is present during the first day of life. It is elicited by positioning the dorsum of the foot against the edge of a table with the infant in vertical suspension. The leg initially flexes (phase 1) and then extends (phase 2), "placing" the foot on the table top. Asymmetry may be indicative of a hemiparesis, asymmetric diplegia, or other neurologic deficit.
The upper-extremity placing response (UPR) is also of great clinical significance. It is elicited by pressing the dorsum of the hand against the edge of a table. The hand is initially lifted above the table top (phase 1), and this is followed by extension of the extremity (phase 2), with "placing" of the hand on the tabletop.
This response is not seen until the infant is three months of age. Many feel that it is associated with a mental age of three months and that - if the neurologic aspects are normal - a nonresponse may be indicative of a mental age below three months.
It is important to realize that in this response the leg or arm must go into an extensor posturing with actual "placing" of the foot (phase 2) on the tabletop. Otherwise, the examiner is just observing an "avoidance" response, which is not to be interpreted as a placing response.
The positive support reflex (PSR) is most important, since a normal response gives a good indication of lower-extremity neurologic integrity. Essentially three findings are possible:
1. Absence of any response. This places the infant in high suspicion of neurologic impairment.
2. Momentary extension, followed by flexion of the lower extremities due to cocontraction of the extensors and flexors of the lower extremities.
3. Full extension, supporting the body.
The neonatal response is elicited by bouncing the hallucal areas of both feet against the tabletop, producing momentary cocontractures of the extensors and flexors of the lower extremities. Thus, the infant momentarily "stands" and then relaxes. This is seen in newborns and becomes more readily elicited in the ensuing two or three months. By six months the PSR fully supports the body in a standing position. If there is noticeable plantar flexion with equinus positioning of the feet, this may be indicative of increased extensor tone and, if unilateral, may be indicative of a hemiparesis or, if bilateral, of a diplegia.
Absence of this reflex is seen in the hypotonic phase of cerebral palsy. This response is also absent in paraplegia caused by deficits in the spinal cord.
Since this particular primitive reflex is somerimes utilized by pediatricians, physical and occupational therapists, and orthopedic surgeons to suggest the need for remedial treatment, without understanding that the treatment will be remedying an obligatory primitive reflex, it may warrant further discussion here.
We grade the positive support reflex by holding the infant in vertical suspension below the axillae, with the head in the midline flexed position (neutral position) and bouncing the child five times on the balls of his feet. The great-toe areas of the feet are then supported on the floor, and the child is held in a vertical position while the physican assesses the degree of supporting response.
Grading the response. We use a five-point scale, ranging from 0 to 4, to quantitate this particular primitive reflex, as follows:
0. Absence of a response. The child remains in flexion, not supporting his weight.
1+. The child momentarily supports his weight (one to 30 seconds).
2+. The child supports his weight for longer than 30 seconds, with a quick movement (under five seconds) from the plantar to dorsiflexion positioning (i.e., the heel makes contact with the examining surface).
3+. There is delay in movement from plantar flexion to dorsiflexion; the child remains in the equinus position longer than five seconds but no longer than 30 seconds).
4+. Obligatory. The child maintains the equinus positioning for more than 30 seconds.
The presence of a 3+ or 4+ response will alert the therapist to the need to prevent contractures. If there is a persistent 3+ response, as seen in the spastic or the child with pyramidal cerebral palsy, then - in addition to physical therapy manipulations - the orthopedic surgeon may place the child in night splints and/or leg bracing during the daytime in order to minimize tendo Achillis contractures.
If an obligatory response (4 + ) is present and the child has a spastic form of cerebral palsy, tendo Achillis lengthening may be indicated. If the child has extrapyramidal cerebral palsy in which this response is variable and contractures are not evident (since the ankles can be brought to neutral position), surgery is contraindica red and night splinrs and/or bracing indicated.
This is the one reflex utilized by orthopedic surgeons in determining whether medical or surgical therapy is indicated, depending on the degree to which this reflex is present.
Many other primitive reflexes are present in infants, but those discussed above are the ones that should be familiar to pediatricians as they examine their patients in and following the neonatal period. Evaluating the findings. When motor development is not progressing in infancy, it is important for the pediatrician to analyze the reason. It may be due to muscle disease or neuropathy when hypotonia is coupled with absent primitive reflexes and deeptendon reflexes. The hypotonia may be of central origin, however, if primitive reflexes are present along with the deep-tendon reflexes.
Infants who have deviant motor development (nonsequential development or motor milestones) might also be placed in high suspicion, especially when a "pseudo" or "splintered" motor milestone is seen. A pronounced primitive reflex at times can produce a motor response that is neither functional nor appropriate for the "motor" age of the infant. For example, the pediatrician should be concerned about an infant who rolls over too early - e.g.; at two months of age. This is usually due to the presence of a significant tonic labyrinthine reflex; the infant assumes an extreme opisthotonic position and, in doing so, flips over. Pediatricians should not be misled by this pseudomotor ability, since it really means that the child is highly suspect of having an abnormal primitive reflex.
The primitive-reflex profile, in addition to the uses described above, can also be helpful in providing objective evidence of cerebral insult. The results of our studies may eventually be able to show a correlation with the minimal brain-dysfunction syndrome, which is frequently found in children with learning disabilities.
We believe it is essential for pediatricians to become knowledgeable about primitive reflexes, since in our opinion they are the most valuable clinical tool for the early identification of motor impairment in infancy.
The research on which this article reports was supported in part by Project 917. Maternal and Child Health Services, and by Maternal and Child Health Services Research Grant MC-R 240392-01-0. US Department of Health. Education, and Welfare.
1. Yang, D. C, Ying, R. Y., and Kennedy, C. The predictive value of the neurologic examination in infancy for mental status at four years of age. In MacKeith, R., and Bax, M. (eds). Studies in Infancy. !Clinics m Developmental Mediane 5er. No. 27). London: The National Spastics Society Medical Education and Information Unit in association with William Heinemann Medical Books, 1968, pp. 94-99.
2. Taft, L. T., and Cohen, H. J. Neonatal and infant reflexology. In Helmuth, J. (ed). Exceptional Infant. Volume 1. New York: Brunner/Mazel, 1967.
3. Capute, A. J., Accardo, P. J., Palmer, F. B., et al. Primitive Reflex Profile: Early Motor Diagnosis. Baltimore: John F. Kennedy Institute, MCH. Grant # MC-R-240392-Ol-O. Five-year study, 1976-1981.
4. Capute, A. )., and Accardo, P. J. Primitive Reflex Profile. Baltimore: University Park Press, 1978.
5. MacKeith, R. C. The placing response and primary walking. Guy's Hosp. Rep. 79 (1965). 294-399.