THE INFLUENCE O F SEX AND AGE ON T H E POSTURAL SWAY O F MAN* FRANCES A. HELLEBRANDT AND G E N E V I E V E L. BRAUN Department of Pliysiology, University of Wismnsin, Madison ONE FIQWE Standing is a complex neuromuscular act, dependent upon the integrity of the myotatic reflex (Sherrington, '24) and much affected by the influence upon the final common path of impulses descending over extra-pyramidal tracts (Magnus, '26). The ever-present collapsing stresses of gravity must be constantly equilibrated by muscular contraction. This involves the harmonious functioning of a large number of muscles acting upon a multijointed structure of a size and proportions which give to the biped stance considerable mechanical instability. It is not surprising therefore that the steadiness of standing has been used as a criterion of both motor power (Bolton, '03) and neuromuscular control (Miles, '22). In 1893 Romberg directed attention to the diagnostic significance of the inordinate augmentation of postural sway evoked in subjects with posterior column disease by simple closure of the eyes and narrowing of the base of support. Efforts had already been made to obtain quantitative measures of the insecurity of the vertical stance (Mitchell and Lewis, 1886). At the instigation of Wier Mitchell, Hinsdale (1887) developed one of the earliest precision devices. This, the atauiograph of Dana (1892) and the ataxiometer of Miles ( '22) evaluate static equilibrium by modifications of a method generally credited to Vierordt, that of measuring postural instability by recording head sway. The method assumes that the multijointed body oscillates in toto over the ankle joint. * Supported in part I)? a grant from the Wisconsin Alumni Research Foundation. 34i AM!2RIPAN JOURV.4L O F PHYSICAL ANTH&OPOU)OY, VOL. XXIV, NO. JANUARY-XARCH. 1939 3 348 F. A. HELLEBRANDT AND G. L. BRAUN I n its application to heterogenous groups a correction for variations in height must be applied. Recently Moss ('31) devised a type of apparatus in the form of an unstable platform which moves in a vertical direction upon a central pivotal point as the weight of the subject shifts from foot to foot and heel to toe. Methods developed in our laboratory (Kelso and Hellebrandt, '37) have made it possible to obtain accurate simultaneous determinations of the shifts in the center of gravity occurring in the two cardinal vertical orientation planes of the body during natural comfortable standing and to project these into the base of support a s a function of time. The object of the study here reported was to determine the influence of sex and age on static equilibrium by this method which evaluates postural stability in terms of one of the most fundamental mechanical factors by which it is conditioned. METHODS By equating moments the maximal shifts in the center of gravity existing simultaneously in the coronal and sagittal planes were estimated from kymographically recorded changes in balance weight which occurred when a subject stood on the uppermost of a series of platforms constructed to affect concurrently two scales placed at right angles to each other. The scales were equipped with levers for the graphic registration of changes in spring tension. A secondary coil was suspended from each lever so as to hang freely in the air-gap of a slotted core transformer with a uniform flux density. The current induced was led to integrating watt-hour meters. From these could be obtained the average position of the lever in any unit of time, hence the average scale load, and by equating moments, the mean location of the perpetually shifting center of weight. The points of maximal sway in the two vertical orientation planes and the mean position of the center of weight during the total period of standing were projected into the base of support which was delimited in the following way. The soles of the feet were moistened with STATIC EQUILIBRIUM 349 potassium permanganate and the subjpct stood on absorbent paper in a fixed position on the platform of the center of gravity apparatus. This left a durable footprint a known distance from the supporting knife edges. The heels were uniformly separated and the toes turned out so that the breadth of the base of support a t the level of the fifth metatarso-phalangeal joints equalled the anteroposterior diameter a s measured from the heel to the great toe. The guide standardizing such a foot posture was removed from the plntform before making the center of gravity determinations. The subjects then stood for 3 minutes, maintaining a natural and comfortable stance with the gaze fixed on a designated area giving a constant visual stimulus. The subjects numbered 111. They ranged in age from 3 to 86. Being drawn largely from the university community they represent a group superior in nutrition, fitness and bearing. This is particularly true of the young adult and middle aged women, most of whom were professional students or teachers of physical education. I n the age group comparisons, two sisters in the fourth decade were discarded because, although manifesting no neurological disease, they demonstrated degrees of postural instability several hundred per cent in excess of the mean value for the remaining. Their inclusion distorted the group findings. Both assumed an extremely lax type of standing which, for most subjects, was impossible to simulate. This left a group of 109, 43 males and 66 females. They were divided into five categories : those of the first decade representing childhood; the second decade, the period of adolescence ; the third decade, young adult life ; the fourth and fifth decades, maturity. The remainder, most of whom were well over 50 were classified as aged. After such condensation the groups were too small to warrant a detailed statistical analysis of the data. 350 F. A. HELLEBRANDT AND G . L. B R A U B RESULTS AND THEIR INTERPRETATION 1. T h e proximity of the center of gravity projection to the geometric cen.ter of the base of support in standing. The area of functional support in the foot position selected forms a trapezoid which may be bounded roughly by tangents to the heel and lateral borders of each foot and a line through the imprint left by the heads of the first metatarsals, since according to Morton ('35) the toes do not participate in normal standing. The geometric center of the trapezoid representing the functional base was introduced into each footprint. The separation between this and the plumb falling from the experimentally determined center of weight was evaluated in terms of eccentricity ratios. The perpendicular distance from the gravity line to the transverse median was divided by one-half the length of the anteroposterior median. Similarly, the perpendicular distance from the projection point to the anteroposterior median was divided by one-half the length of the lateral median. The eccentricity was trifling in extent but significantly consistent in its direction. In 85% of the subjects the projection of the center of weight fell slightly in front of the geometric center of the base. The mean eccentricity ratio in the anteroposterior plane was f0.2309. I n only 15% of the subjects did the experimentally determined center fall behind the mathematically derived one, and the mean ratio was -0.0541. Thus we see that not only is there a tendency for the average center of weight to be held in front of the geometric center of the base, but the eccentricitv in this direction is 76% greater than the deviation backward. Basler ('29) reported that when his subject assumed a free standing position with the feet parallel, the weight l h e cut the base of support about 4 cm. to the right of the midline, within the area covered by the right foot. Reynolds and Hooton ('36) also noted a lack of symmetry in the stance of their subjects. I n about one-third the weight line coincided with the middle of the base. I n one-third it fell to the right and in the remaining to the left. The asymmetry was unrelated to inequality in leg length. With a method slightly STATIC EQUILIBRIUM 351 more accurate than that of Reynolds and Hooten but identical with it in principle we found the gravity line equidistant between the heels in only 10.98% of 327 college girls standing in their best posture and 13.25% of eighty-seven professional physical education students assuming a comfortable stance (Hellebrandt et al., ,'38). The weight fell preponderantly to the left in the natural stance but when a deliberate effort was made to assume a good posture, the center of gravity drifted to the right. The data of all of these investigations (Basler, Reynolds and Hooten, and Hellebrandt) came from single instantaneous observations of the center of gravity. The findings of the present study have considerably more significance because each observation represents the mean position of the center of weight during 3 minutes of standing. The location of the center of gravity of the normal subject in the vertical posture oscillates rhythmically and the extent and range of movement may sometimes be extreme (Hellebrandt, '38). I n 72% of the subjects of the present study the mean eccentricity was to the left, averaging -0.1051. I n the remaining 28% it was to the right, the mean value being +0.0838. The center of weight deviates more frequently to the left, confirming our original observation on natural standing, and exceeds by 20% the eccentricity to the right. In 63% the eccentricity was to the left and in front of the geometric center of the base; in 22% to the right and in front; in 9% behind and to the left; and in 6% behind and to the right. The peculiarities in the eccentricity of the center of weight in its relation to the geometric center of the base were borne out in both sexes. There were no significance age differences. Hinsdale (1887) believed that his subjects swayed to the heavier and stronger side, forward and to the right. Our subjects are falling away from the heavier side, if the right is more developed than the left as is generally accepted. They appear to over compensate for their right-sided dominance. The tendency to overshoot in the forward direction is more easily explained. The center of gravity falls considerably in front of the transverse axis of the ankle joint 332 F. A. HELLEBRANDT AND Q. L. BRAUN thus creating a rotatory stress which tends constantly to project the body forward. It must, however, be emphasized that the mean eccentricity of the center of gravity projection in both of the cardinal vertical orientation planes is in general very slight. Haycraft (Schafer, ’00)has reviewed the early, conflicting evidence which variously placed the plumb passing through the center of gravity in front of, above or behind the joints of the weight bearing limbs. Steindler (’35) fixes the gravity line “about 4 cm. in front of the center of the ankle joint.” We observed it 4.92 cm. anterior to the malleolus in 327 college freshmen standing in their best posture, and 5.65 cm. in 87 professional physical education students maintaining a natural comfortable stance (Hellebrandt et al., ’38). The difference was more apparent than real because of group dissimilarity in average foot length. We have already called attention to the fallacy of relating the gravity line to the ankle joint without heed to the diameters of the base. Our present findings indicate that the perpendicular dropped from the average location of the center of weight falls remarkably close to the geometric center of the base. This is in keeping with the dictates of stable construction and is apparently a fundamental relation not easily altered. We have demonstrated that it remains undisturbed by the heel height of the shoes worn by women unless their pitch is extreme (Hellebrandt et al., ’37). Compensation is automatically achieved by knee flexion and lordosis. With each new base there is a new posture, and an accurate preservation of equal stability in all diameters of the base, the center of weight projection always occupying a very nearly middle position within the area of support. 2. T h e magnitude of the postural instability. Postural sway was slight when measured in terms of shifts of the center of gravity. The maximum displacement of the center of weight in the two vertical orientation planes was projected into the base. When the area enclosed by these four points indicating greatest sway forward, back, to the right and left, was estimated as a percentage of the total area of support, it amounted STATIC EQUILIBRIUM 353 on the average to 1.28%, ranging from 0.15 to 6.13 in the steadiest and the most unstable respectively. I n absolute units, the area of maximum sway ranged from 0.5 to 13.7 sq.cm., averaging 3.73. Small as is the area of underpropping in relation to the disproportionate height of the center of gravity of the human body in the vertical posture, it includes a relatively enormous margin of safety. The involuntary shifts in the center of weight during standing occupy a very small core surrounding the geometric c h t e r of the base. The area of maximum sway in the two unstable subjects who were discarded from the group findings measured 19.9 and 26.9 sq.cm. Even these exaggerated degrees of instability encroached respectively upon only 5.41 and 7.35% of the base. The group studied was anything but homogeneous. The subjects varied in age from 3 to 86 and differed widely in height, weight and configuration. Yet the eccentricity of the mean location of the center of gravity projection was singularly minute, and the base exceeded that demanded because of involuntary postural swap by a notably generous margin. I n spite of mechanical factors in build conducive to static insecurity, the normal subject maintains the biped stance with an inordinate degree of symmetry and stability. These observations suggest, on a priori reasoning, that this is an inherent phenomenon under reflex control. The effective stimulus for the postural contraction of the antigravity muscles is stretch (Liddell and Sherrington, '24). The receptors involved are the proprioceptive end-organs of the muscles primarily concerned. If the myotatic reflexes are impaired, as they are in posterior column disease, standing is almost impossible, particularly when the eyes are closed. Tonus is much affected by visual, auditory, labyrinthine, touch and pressure sensations. 3. Age differences in postural sway. Balance, as demonstrated by a mean centering of the weight over the middle of the base, seems from our data to be uniformly good at all ages. The magnitude of the oscillation of the center of gravity about the geometric center of the base is, however, less constant. Figure 1illustrates the tendency for the young and the 354 F. A. HELLEBRANDT A N D G. L. BRAWN 9 li Fig. 1 Graphs showing the relation of postural stability as measured by shifts in the center of gmvitr of the body t o age and to certain characteristics of the base of support. STATIC EQUILIBRIUM 355 old to be less stable than the young adult and the middle aged subject of both sexes. Hinsdale (1887) and Hancock (1894) had observed that children sway more than adults. In our experiments the area enclosed by the maximal shifts of the center of gravity in the cardinal vertical orientation planes and its relative encroachment upon the limits of static security fall and then rise when graphed against age. The differences however, may not be statistically significant. Many of the subjects in the very stable young adult group were professional students of physical education possessed of a better than average neuromuscular development. Nevertheless, morphological evidence is in agreement with the expectation of maximal postural stability in the third decade. Many years ago Flechsig suggested that the development of the nervous system could be traced by ascertaining periods at which various fiber tracts acquire myelin sheaths. I n general the acquisition of myelin occurs in the same order as that in which nerves develop. Medullation is demonstrable early in fiber systems which are phylogenetically old and in those becoming functional first. The posterior columns, spinocerebellar and cortico-spinal pathways are among the earliest fiber tracts to undergo myelination (Coghill, '29 ; Langworthy, '26, '28). Myelin first appears in the human in the column of Burdach at about 14 weeks (Keene and Hewer, '31; Arnott, '37). It is demonstrable in the spino-cerebellar tracts by the sixteenth week. The afferent limb of the long circuit myotatic reflexes thus originates at a time which antidates any utility. The efferent limb, pyramidal and extra-pyramidal, matures much later. Myelination of the pyramidal and rubro-spinal tracts is scant at term and is not complete in the former until the second year. Langworthy ( '33) believes that myelination continues until the beginning of puberty. Hand in hand with this orderly growth of the nervous system there is a similar uniformity as to the time of appearance of motor behavior patterns (Abramson, '37). Maturation rather than exercise seems to play the dominant role in the initiation of basic motor 356 F. A. HELLEBRANDT AND 0. L. BRAUN acts and the sequence of the development of fiber tracts from which behavior patterns are formed is determined by evolutionary influences (Herrick, '30). It is impossible to say, however, from the scattered evidence at hand, if the growing child is at any significant neurological disadvantage in the maintenance of a neuromuscular act as complex as the biped stance. I n 1936 Newman and Corbin reported an increase in the threshold of vibratory acuity with age, the loss being distiiict at 30 and very striking in those beyond 50. The vibratory sense was most acute in the second and third decades. Subsequently Corbin and Gardner ('37) counted the number of myelinated fibers in the dorsal and ventral roots af cadavers varying in age from 1 day to 84 years. They found a degeneration of peripheral process from dying neurones in the spinal ganglia which they believed might account for the progressive loss of vibratory sense with age. This suggests that there might be similar degeneration in the aged of posterior column afferents mediating proprioceptive impulses. The crowning instability was present in the very young. This may have causes other than the neurological one suggested. Mechanical factors in build probably play some part. The physical proportions of the child are not conducive to postural stability. The lower extremities are ill developed and the center of gravity of the body as a whole has been found by Basler ('35) to be relatively higher in the younger age groups of girls. He found, however, no such relationship to age in boys. The age differences in the various foot length diameters of Morton and the area enclosed between the weight bearing imprints of the base of support are presented in figure 1. From the second decade on the base remains more or less uniform in size. Inadequacy in muscular development may also account in part f o r the instability of the young. Peak muscle power is reached relatively late and it falls off in the aged (Bolton, '03; Lipovetz, '33). This is, however, probably of minor importance because the equilibrium of collapsing gravital STATIC EQUILIBRIUM 357 stresses is met very economically by the postural contraction of the anti-gravity muscles (Simonson, ’26 ; Tepper and Hellebrandt, ’38). The energy metabolism is little increased by the vertical in contrast with the horizontal posture, even when instability is great, involving the participation of a large number of skeletal muscles. We have observed a case of muscular dystrophy in a boy of 14 who reported gradually increasing difficulty in walking which became seriously incapacitating during the winter because not enough strength remained to lift the feet in the snow. The boy was unable to get up from the sitting position without assistance and had to be lifted onto the platform of the center of gravity apparatus. Yet he stood perfectly, with little involuntary postural sway and the Romberg sign was negative. A similar observation was made on a case of Addison’s Disease followed through several crises. Although muscular weakness was profound, standing was essentially normal when measured in terms of the ability to retain the center of weight in the middle of the base with a minimum of oscillation. Suggestive as these case histories are, we have no direct evidence that the antigravity muscles of the weight bearing limbs are sufficiently developed during the first decade of life to meet the demands placed upon them by the assumption of a vertical stance as effectively as they are met in later life. To stand quietly in a natural, comfortable posture for even as short a time as 3 minutes is a difficult feat for the young child. Every extraneous motion is reflected by a change in the location of the center of gravity unless exactly counterbalanced. I n the handling of the older group the converse problem presented itself. Few subjects know how to relax while maintaining a vertical stance. The habit of ‘good posture,’ when strongly developed, introdnces an element of cerebral interference which cannot be broken down without prolonged training. It was sometimes difficult to know when great steadiness was an artefact of this origin in the young adult and when instability mas a manifestation of restlessness in the young. 355 F. A. HELLEBRANDT AND 0. L. B U U N Hancock (1894) observed that girls were steadier than boys. The mean area of maximal sway was 1.06 sq.cm. in our total group of sixty-six females, occupying 3.28% of a base which averaged 433.8 sq.cm. in contrast with 1.59 sq.cm. in our group of forty-three males. Although the mean base in this sex is larger, 462.1 sq.cm., the area of maximal sway occupied 4.4% of the total base, codrming Rancock’s observation. SUMMARY AND CONCLUSIONS Postural stability was measured in 109 subjects, 43 male and 66 female, ranging in age from 3 to 86, using as a criterion, shifts in the center of gravity of the body as a whole projected into the base of support as a function of time. The following conclusions may be drawn from the evidence presented: 1. A t all ages and in both sexes the mean position of the center of weight during 3 minutes of standing falls remarkbbly close to the geometric center of the base. 2. The mean eccentricity of the center of gravity is slight, falling preponderantly in front and to the left of the geometric center of the base. 3. The area enclosed by the maximal shifts of the center of gravity in the cardinal vertical orientation planes is very small, averaging 3.73 sq.cm. or 1.28%of the total base. 4. The foot posture of Morton affords a base of support in the vertical stance with a relatively enormous margin of safety. 5. The tendency is for postural stability to be greater in the young adult than in the child or the aged, but variations in muscular power, neurological development, factors in build and disturbances of psychic origin give the data presented no quantitative significance. 359 STATIC EQUILIBRIUM LITERATURE CITED ABUMSON,HAROLD 1937 Graded sequence i n postural and in locomotor development. Am. J. Dis. Child., LIII, 1282. ARNOTP,D. W. H. 1937 A study of myelination and how it helps in understanding the influence of stimulus on the development of the nervous system. Med. J. Australia, I, 611. BASLEB, ADOLF 1929 Die Schwerlinie des menschlichen Korpers uud ilire Verschiebungen. Pfliiger 's Arch. ges. Physiol., CCI, 768. 1935 Uber die relative Schwerpunkthohe bei Knaben und Wadchen. Z. Mensch. Vererbungs-u. Konstitionslehre, IXX, 90. BOLTON,T. L. 1903 The relation of motor power to intelligence. Am. J. Psychol., XIV, 351. COOHILL,G. E. 1929 Anatomy and the Problems of Behavior. New York. C O ~ I N KENDAL , B., A N D ERNESTD. GARDNER 1937 Decrease in number of mpelinated 5bers in human spinal roots with age. Anat. Rec., LXVTII, 63. DANA, CHARLES L. 1892 Text-Book of Nervous Diseases. New York. HANCOCK,JOHN A. 1894 A preliminary study of motor ability. Pedagogical Seminary, 111, 9. HELLEBRANDT, FUNCES A., DORIS RUBIN, WINIFRED &f. LONGFIELD A N D L. E. A. KELSO 1937 The base of support in stance. Physiotherapy Rev., XVII, 231. HELLEBRANDT, FRANCES A., RWYEI€.'I'EPPER, GENEVIEVE L. BRAUNAND MARGARET C. ELLIOTT 1938 The location of the cardinal anatomical orientation planes passing through the center of weight in young adult women. Am. J. Phyeiol., CXXI, 465. HERRICK,C. JUDSON 1930 Localization of function in the nervous system. Proc. Nat. Acad. Sci., XVI, 643. HINSDALE, GVY 1887 The station of man considered physiologically and clinically. Am. J. Yed. Sc., XCIII, 478. KEENE,M. F. L u u , AND E. E. HEWER1931 Some observations on myelination iii the human central nervous system. J. Anat., LXVI, 1. 1937 KELSO, L. E. A., AND F. A. H E L L E B R A ~ Devices for the study of two plane shifts in the center of gravity of a swaying body. Science, LXXXVI, 451. LANGWORTHY, ORTHELLOR. 1926 Relation of onset of deeerebrate rigidity to time of niyelinization of tracts in the brain-stem and spinal cord of young animals. Contrib. Embryol., no. 89, Carnegie Inst. Wash. 1928 The behavior of pouch-young opossums correlated with the myelinization of tracts in the nervous system. J. Comp. Neur., XLVI, 201. 1933 Development of behavior patterns and myelinization of the nervous system in the human fetus aad infant. Contrib. Embryol. no. 139, Carnegie Inst. Wash. LIDDELL, E. G. T., AND CHARLES 8. SHEZEINGTON 1924 Reflexes in response t o stretch. Proc. Roy. Soc. London (B), XCVI, 212. LIPOVETZ, Fau, JOHN 1933 Physiology of Exercise. Minneapolis. MAGNUS,R. 1926 Physiology of posture. Lancet, 11, 531, 685. AMERICAN J O U R N A L Or PHYSICAL ANTHBQPOIIMY. VOL. XXIV, NO. 3 360 F. A. HELLEBRANDT AND G. L. B R A U S MITCHELL, S. WEIR, AND MORRISJ. LEYWIS1886 The tendon-jerk and musclejerk in disease and especially in posterior sclerosis. Am. J. Med. Sc., XCII, 363. MILES, W. R. 1922 Static equilibrium as a useful test of motor control. J. Indust. Hygiene, 111, 310. MORTON, DUDLEY JOY1935 The Human Foot. New York. Moss, F. A. 1931 Development of the Wabblemeter for measuring bodily steadiness as an indication of riding comfort. J. Soc. Automotive Engineers, X X V I I I , 580. N E ~ Y A NHENTLY , W., AND KENDALL B. COBBIN 1936 Quantitative determination of vibratory senaibility. Roc. SOLExp. Biol. and Med., XXXV, 273. RBYNOLDE,EDWARD AND E. A. HOOTON 1936 Belation of the pelvis to erect posture. Am. J. Phps. Anthrop., XXI, 253. ROMBEXG,M o a r n HEINRICH 1893 A Manual of the Nervous Diseases of Man. London. SCHAFXX, E. A. 1900 Text-Book of Physiology. Edinburgh. SHERRINOTON, SIR CEARLEX 1924 Problems of muscular receptivity. Nature, CXIII, 892, 929. SIYONSON,ERNST 1926 Zur Physiologie des Energieumsatzes beim Menachen ; zur Physiologie des Stehens. Arch. ges. Physiol., CCXIV, 403. STDXNDLER, ARTHUR 1935 Mechanics of Normal aud Patliological Locomotion i n Man. Baltimore. 1938 The influence of the TGPPE&,RUBYN H.,AND FRANCES A. H~LEBRANDT upright posture ou the metabolic rate. Am. J. Physiol., CX S I , 563.