Relationship of turnout to hip professional ballet dancers S. KUSHNER,*† MSc
From the
*
Glen Sather
(PT),
L. T.
abduction in
SABOE,‡ BPT, D. REID,*‡ MD, FRCS(C), PENROSE,* MSc, AND M. GRACE,‡ PhD
Sports Medicine Clinic and the ‡ Division of Orthopaedic Surgery, the University of Alberta, Edmonton, Alberta, Canada visible joint on stage, controlling the position of the whole lower limb.17 With turnout, the lower extremity profile is presented to the audience, allowing the dancer to demonstrate body motion, strength, and beauty.14, 21 Ideally, the turnout comes from external rotation of the femur. In weightbearing, many dancers compensate for insufficient hip motion by rotation at the knees, eversion of the heels, pronation of the feet, and lordosis of the lumbar spine.6, 9, 14 Thus, it is possible that with many dancers, the great emphasis on turnout may be the origin of some of the spinal and lower extremity injuries seen in ballet. External rotation is not only aesthetically desirable but is believed by dancers to be critical to achieve maximal abduction of the hip. Abduction is limited by impingement of the greater trochanter on the superior rim of the acetabulum and adjacent ilium if performed without rotation of the hip. With increasing external rotation, the pelvis tilts to the opposite side and the head of the femur assumps a more inferior position, delaying contact and hence allowing a greater range of abduction (Fig. 2). External rotation of the hip is essential to good ballet technique. Insufficient range of turnout or faulty techniques may predispose dancers to injury. Dance masters’ traditional emphasis on prevention of injuries through fault correction and increasing external rotation is probably a sound principle. However, is a less extreme turnout sufficient for this
ABSTRACT The ability to externally rotate or turn out the hip is fundamental to ballet. Every classical dancer aims to achieve perfect turnout. The purpose of this study was to determine how much turnout is necessary for maximal abduction. It was hypothesized that moderate turnout is sufficient for this purpose. Twenty-two professional dancers from the Alberta Ballet Company were studied. Measurements of passive hip abduction were taken at 0°, 45°, 60°, 70°, 80°, 90° and maximum hip lateral rotation using a goniometer and Leighton
flexometer. Statistical analysis was done using Pearson correlation coefficients. A significant positive correlation was found between abduction and lateral rotation ( P< 0.05). The greater the position of external rotation, the more abduction achieved. In conclusion, the traditional emphasis on good turnout has some scientific merit and functional implications.
The ability to externally rotate the hip is fundamental to all forms of dance, especially ballet. In ballet, steps and movements proceed and terminate from five basic positions of the feet that are largely determined by external hip rotation or turnout (Fig. 1).6, 12 In the first position, for example, the hips are usually forced to their limits of external rotation.&dquo; Every classical dancer aims to achieve perfect turnout, which is defined as &dquo;the ability of a dancer to turn his or her feet and legs out from the hip joint to a 90° position. This turnout is one of the essential principles of classical dance, giving the dancer freedom of movement in every direction.&dquo;5 Together with the pelvis, the hip is the focal point of the dancer’s concentration. It is perhaps the most
purpose? Previous publications have described problems associated with insufficient hip motion, sometimes compounded by a late start in ballet with subsequent inability to alter hip anteversion or ligamentous structures.2, 3, 13,15 However, we were unable to find any studies done to determine the ideal functional hip range for ballet. Specifically, there is no information on the point at which additional external hip rotation effectively increases the amount of obtainable hip abduction. We investigated this relationship in this study, hypothesizing that the traditional emphasis on extreme
t Address correspondence and repnnt requests to’ Shirley Kushner, Glen Sather Umversity of Alberta Sports Medicine Clinic, E-05 Van Vliet Centre, Edmonton, Alberta, T6G 2H2 Canada. 286
287
turnout may not be necessary to obtain maximum abduction of the hip.
METHODS As part of preseason medical screening, all members of the Alberta Ballet Company were studied. Twenty-two professional dancers (8 males, 14 females) from the company had bilateral hip abduction measured with increasing amounts of external rotation. After testing procedures were explained to the dancers, they performed an informal, traditional warmup. Testing began when the dancer felt that maximal flexibility had been obtained. Goniometric measurements were taken with the dancer in the supine position. Pelvic stabilization was achieved by hanging the contralateral leg over the table while the anterior superior iliac spine was held down (Fig. 3). Abduction was measured with a long arm goniometer (Goniometer, Fred Sammons Inc., Brookfield, IL), the axis of which was placed over the center of the hip joint. The center was determined by palpating the femoral artery immediately distal to the inguinal ligament. The fixed arm was aligned parallel with the body midline, while the moving arm was aligned with the long axis of the thigh between the centers of the hip and knee joints. Rotation was controlled by a second investigator with a Leighton flexometer (Leighton Flexometer, Spokane, WA) strapped to the sole of the dancer’s foot, which was held at 0° of dorsiflexion. The Leighton flexometer is a goniometer comprised of a weighted 360° scale and a weighted pointer that move independently of each other. This device was used because of its high degree of reliability.7, 10 The axis of motion was centered over the second metatarsal head so that in neutral hip rotation the foot was perpendicular to the floor
Figure 1. Five basic foot positions in ballet.
was locked at zero, with the pointer indicating 0°. The second investigator then maintained the selected external rotation position by holding the calcaneus and maintaining the subtalar joint in a neutral position while the knee was in extension. The test positions were in sequence : 0°, 45°, 60°, 70°, 80°, 90°, and maximal rotation. In each of these positions the hip was fully passively abducted. Rotation was rechecked and the amount of abduction recorded with the long goniometer arm for each of the seven test positions. Hip flexion was prevented until the position of maximal external rotation was reached. At this point, approximately 20° of flexion was allowed, since without this flexion, the dancers found it impossible to achieve maximal abduction. Both hips were tested, but right and left hips were measured in a random order.
and the scale
RESULTS The mean age of the dancers was 19.4 years for females and 24 for males. They had 12.7 and 8.5 years of training,
respectively (Table 1). None of the dancers had a history of hip pain severe enough to require either medical treatment or a day off dancing in the 6 months before testing. Analysis was done with t-tests and Pearson correlation coefficients using a 0.05 level of significance. The t-tests were performed in order to determine whether the male and female dancers differed significantly. There was a significant difference between males and females for age, height, weight,
and years in ballet (Table 1). Dancers were classified according to dominant and nondominant sides as determined by right and left handedness. All but two of the females and one of the males were right-handed. There was no significant difference between the sexes for amount of abduction achieved with the dominant or nondominant leg at 0°, 45°, 60°, 70°, or 80° of external rotation (Table 2). There was a significant difference (P < 0.05) between the sexes at 90° of external rotation for abduction achieved with the dominant limb, with greater abduction shown by the female group. There was a tendency toward significance in the nondominant limb. Pearson correlation coefficients were used to determine association between lateral rotation and abduction. A regression line was fitted showing the abduction values obtained at each level of rotation. This line of best fit was linear and there was a significant positive correlation between abduction and lateral rotation (P < 0.05). As the lateral rotation increased, degrees of abduction also increased (Fig. 4). Both males and females were able to attain more abduction in their nondominant limbs than in the dominant ones. There was a significant difference (P < 0.05) in abduction values between rotation at 0° with each of the rotation values at 45°, 60°, 70°, 80° and 90° as analyzed by paired t-tests.
288
Figure 2. Position of the pelvis and the head of the femur (Adapted from Sparger.19)
in the acetabulum
DISCUSSION Classical ballet emphasizes the importance of turnout for both aesthetic reasons and functional considerations. This study verifies the positive correlation between external rotation and the ability to achieve a greater range of abduction. The empiric wisdom of dance masters in encouraging external rotation is rewarded by facilitating abduction. Increased abduction values were present in the nondominant limbs of both male and female dancers (Fig. 4). This may be due to tighter and possibly stronger extremities on the dominant sides. The amount of abduction in this study was greater in females than in the male dancers for all of the external rotation values (Table 2). The remarkable flexibility
during (A) early
abduction and
(B)
late abduction.
achieved by dancers is linked in part to the age at which dance training is initiated. In this study of professional dancers, the females began their training on average at 6.6 years and the men at the age of 15.4 years. These ages correspond to data recorded in other series. 13 The fact that males started serious dancing during adolescence is probably one of the factors contributing toward their inability to achieve the same external rotation and hip abduction values as the females.&dquo; Miller et al. 13 have reported that male ballet dancers suffer more knee pain than females. This may be due in part to lack of hip external rotation and subsequent greater stresses on the medial joint line of the knee. Another contributing factor might be the more frequent and larger jumps, more complicated smaller jumps, and the superimposed stresses
289
Figure 3. Test position with the contralateral limb hanging over the table edge to assist in pelvic stabilization in (A) early abduction, (B) late abduction, and (C) following the 20° of flexion allowed. TABLE 1
Physical characteristics of male and female
dancers
TABLE 2
Mean hip abduction at various degrees of external rotation
Figure 4. As the external rotation increases, abduction.
so
does the
290
hip contribution, accounting for 70%
of turnout in beginners
and 68% in advanced dancers. These authors concluded that since the ideal alignment is not totally achievable at the hip, the turnout may result in significant stresses at the knee, ankle, and foot. There may be some relationship between the increasing contribution to turnout by the knee, foot, and ankle and increasing injury in these areas.11, 15, 16 In this study, although we attempted to limit the turnout of the hip by having the subjects maintain full knee extension and control foot position, some movement probably occurred at
5. The main determinants of turnout. Hardaker et al .6)
Figure
(Adapted from
TABLE 3
Range of abduction available with maximum external rotation (in degrees)
of lifting in the male dance role. The ligamentous structures of younger individuals are more adaptable. Sammarcol7 has reported that children who start dancing between 6 and 12 years of age acquire their turnout during the period when femoral neck-shaft relationships are maturing and that, after 11 years, the shape of the femoral neck cannot be altered. Retroversion, thus, cannot be increased.&dquo; Dancers who start after 11 years of age probably increase flexibility by stretching the hip capsule.2 Miller et al.l3 have reported that in the adult dancer, extraordinary amounts of external rotation have not been based on femorai neck retroversion when tested radiologically. If one concedes that hip range is acquired by task-related activities, it is interesting to review the factors that potentially limit or predispose to hip motion (Fig. 5). The extent of the turnout is limited by: 1) the femoral anteversion angle-the greater the anteversion, the less the range achievable ; 2) capsular flexibility of the hip, which, in turn, is related to generalized ligament laxity and collagen type, 3) the orientation of the acetabulum-the more anterior the inclination, the earlier the impingement of the femoral neck on the posterior lip of the acetabulum, and 4) the flexibility of the muscles and tendons crossing the hip joint.6, 12 Obviously, some of these factors are subject to training influence, whereas others are fixed. While 60° to 70° of turnout occur at the hip, some 20° to 30° may be accommodated by knee and foot alignment.’ Meinel and Atwaterll have reported that 55% of beginners and 52% of advanced or experienced dancers’ straight-legged turnout occurs at the hip. In the demi-plie, there is a greater
these joints. The dancers in our study group (males and females considered respectively) were homogenous with very little variation in height, weight, age, or hip abduction at each of the rotation values. This may support claims that dancers’ flexibility is a reflection of candidate selection in which individuals who are more flexible may become elite dancers.’ Hardaker et al.6 describe a &dquo;Darwinism of dance&dquo; whereby the professional dancer, like the elite athlete, represents the successful apex of a large field of dance students, most of whom have failed to pursue their careers because of anatomical variations such as decreased flexibility, body type, femoral anteversion, genu valgum or varum, leg length discrepancy, rigid pes planus, and pes cavus. Many dancers reported episodes of a locking or catching sensation in their hips during abduction of the free limb while dancing. Although we did not attempt to examine turnout
during weightbearing, impingement type pain
was
recorded at the limit of the range of passive abduction. During testing, three of the females and one male reported acute sharp pain during passive motion when at the limit of passive abduction in 70° to 90° of rotation. This pain was so severe that they immediately internally rotated their hips, freeing their legs from the second examiner’s hand. The pain was instantly resolved and the dancers described this mechanism and pain similar to that actively acquired when abduction was performed during dance. The pain may have been caused hv the ilinn-nnq tpnrlnn slipping nvPr an n«eous ridge on the lesser trochanter or the pectineal eminence or from the iliofemoral ligament snapping over the femoral head.8, 18 At the limits of external rotation and abduction, some dancers wanted to assume the slightly flexed position; this had to be actively controlled by the examiners. In dance, abduction movements would be done in slight hip flexion rather than in a purely frontal plane. Therefore, up to 20° of flexion was permitted so that maximal abduction could be obtained in a more realistic dance position (Fig. 3, B and
C). In situations where dancers apparently could abduct al180°, we believe that the movement was actually achieved by a combination of abduction and external rotation followed by flexion with internal rotation to return the limb to neutral (Fig. 3C). This motion is similar in many ways to Codman’s paradox at the shoulder, where rotation, abduction, and flexion maneuvers are coupled as consequential movements that bring about an involuntary movement of spin about the longitudinal axis of the humerus.I most
291
The sequence of events at the hip includes abduction with increasing external rotation to avoid contact of the greater trochanter against the ilium. At maximum abduction, there is tightness of the capsular structures and possibly impingement of the neck on the posterior acetabular lip. Further abduction at the hip can then only be achieved by slight flexion, which removes the contact of the neck because it is linked by consequential motion to internal rotation. This simultaneously relaxes the anterior capsule, permitting the completion of full abduction in slight flexion. Only the extraordinary length of the hamstrings in dancers permits the enormous arc of motion. This explains the apparent neutral position of the foot in figure 3C. The reported maximal rotation values are high and are probably a sum of full external rotation and the internal rotation that occurred as a result of the consequential movement (Table 3). These opposite rotations were probably recorded on the Leighton flexometer as total rotation. This explains why such high external rotation values were recorded (Table 3). These may be erroneous values. We recommend that in future studies the point at which the dancer needs to flex to continue abduction be the point at which maximal external rotation be recorded. We are not reporting on the significance of this data; however, it is interesting to note that a positive correlation was found between abduction achieved and maximal external rotation. A positive correlation was also found between years in ballet and the maximum rotation achieved. We have previously reported that internal rotation and hip adduction is frequently less in dancers than in the normal athletic population. 16 Furthermore, the older and more experienced the dancer, the more this trend is exaggerated and the greater the range of passive hip external rotation possible. 12, 16 We believe that the influence of external rotation on abduction ceases at the point at which flexion has to be allowed to achieve further elevation. For this reason, abduction in a functional dance situation is supplemented by early side flexion of the pelvis and flexion of the hip (Fig. 2). In summary, when testing passive hip elevation through abduction in ballet dancers, the greater the position of external rotation or turnout, the more abduction is achieved. Therefore, we conclude that the traditional emphasis on a
good turnout
has
some
scientific merit and functional im-
plication. ACKNOWLEDGMENTS We would like to thank Sharon Krantz, Ballet Mistress, and Bert Chesworth for their advice and assistance and the Alberta Ballet Company of Edmonton, Alberta for their
cooperation. REFERENCES Basmajian JV, MacConaill MA: Muscles and Movements. A Basis for Human Kinesiology Baltimore, Williams & Wilkins, 1969, p 38 2. Bergfeld JA (Moderator): Medical problems in ballet A round table. Physician Sportsmed 10(3). 98-114, 1982 3. Chmelar RD, Schultz BB, Ruhling RO, et al A physiologic profile comparing levels and styles of female dancers. Physician Sportsmed 16(7) 87-96, 1.
1988 4. Dolinar L. Ballet? MD finds it
rough
as
any sport. Physician Sportsmed 4:
86-92,1976 5 Grant G. Technical Manual and Dictionary of Classical Ballet. Third edition. New York, Dover Publications, 1982, p 122 6 Hardaker WT, Enckson L, Myers M: The dancer as athlete, in Shell C (ed) 1984 Olympic Scientific Congress Proceedings, Vol 8, Champaign, II, Human Kinetics, 1986 7 Hsieh CY, Walker JM, Gillis K: Straight-leg-raising test. Comparison of three instruments. Phys Ther 63: 1429-1433, 1983 8. Jacobs M, Young R: Snapping hip phenomenon among dancers Am Corr Ther J 32 92-98, 1978 9. Lawson J: Teaching Young Dancers—Muscular Coordination in Classical . London, Adam & Charles Black Publishers Ltd, 1984 Ballet 10. Leighton JR: An instrument and technic for the measurement of range of joint motion. Arch Phys Med Rehabil 36. 571-578, 1955 11. Meinel KK, Atwater AE. Analysis of components of the "turnout" in beginning and advanced ballet dancers. Med Sci Sports Exerc 20 (Suppl) 2: S-
6, 1988 12
Micheli LJ, Gillespie WJ, Walaszek A. Physiologic profiles of female professional ballerinas Clin Sports Med 3. 199-208, 1984 13 Miller EH, Schneider HJ, Bronson JL, et al: A new consideration in athletic 181-191, 1975 injuries. The classical ballet dancer Clin Orthop 111: 14. Peterson D: Learning through observation: The dancer’s technique, in Peterson D, Lapenskie G, Taylor A (eds): The Medical Aspects of Dance London, Ontario, Sports Dynamics, 1986, pp 49-62 15 Reid DC: Prevention of hip and knee injuries in ballet dancers. Sports Med 6 295-305, 1988 16. Reid DC, Burnham RS, Saboe LA, et al: Lower extremity flexibility patterns in classical ballet dancers and their correlation to lateral hip and knee injuries Am J Sports Med 15. 347-352, 1987 17. Sammarco J: The dancer’s hip Clin Sports Med 2 485-498, 1983 18 Schaberg JE, Harper MC, Allen WC The snapping hip syndrome. Am J Sports Med 12 361-365, 1984 19 Sparger C: Anatomy and Ballet. London, Adam and Charles Black Publish-
ers Ltd, 1975, pp 24-25 20. Vincent L The Dancers Book of Health. Kansas City, Sheed Andrews & McMeel Inc, 1978, pp 109-112 21. Wheeler L Common musculoskeletal dance injuries. Chiropractic Sports Med 1. 17-23, 1987
From the
*
Glen Sather
(PT),
L. T.
abduction in
SABOE,‡ BPT, D. REID,*‡ MD, FRCS(C), PENROSE,* MSc, AND M. GRACE,‡ PhD
Sports Medicine Clinic and the ‡ Division of Orthopaedic Surgery, the University of Alberta, Edmonton, Alberta, Canada visible joint on stage, controlling the position of the whole lower limb.17 With turnout, the lower extremity profile is presented to the audience, allowing the dancer to demonstrate body motion, strength, and beauty.14, 21 Ideally, the turnout comes from external rotation of the femur. In weightbearing, many dancers compensate for insufficient hip motion by rotation at the knees, eversion of the heels, pronation of the feet, and lordosis of the lumbar spine.6, 9, 14 Thus, it is possible that with many dancers, the great emphasis on turnout may be the origin of some of the spinal and lower extremity injuries seen in ballet. External rotation is not only aesthetically desirable but is believed by dancers to be critical to achieve maximal abduction of the hip. Abduction is limited by impingement of the greater trochanter on the superior rim of the acetabulum and adjacent ilium if performed without rotation of the hip. With increasing external rotation, the pelvis tilts to the opposite side and the head of the femur assumps a more inferior position, delaying contact and hence allowing a greater range of abduction (Fig. 2). External rotation of the hip is essential to good ballet technique. Insufficient range of turnout or faulty techniques may predispose dancers to injury. Dance masters’ traditional emphasis on prevention of injuries through fault correction and increasing external rotation is probably a sound principle. However, is a less extreme turnout sufficient for this
ABSTRACT The ability to externally rotate or turn out the hip is fundamental to ballet. Every classical dancer aims to achieve perfect turnout. The purpose of this study was to determine how much turnout is necessary for maximal abduction. It was hypothesized that moderate turnout is sufficient for this purpose. Twenty-two professional dancers from the Alberta Ballet Company were studied. Measurements of passive hip abduction were taken at 0°, 45°, 60°, 70°, 80°, 90° and maximum hip lateral rotation using a goniometer and Leighton
flexometer. Statistical analysis was done using Pearson correlation coefficients. A significant positive correlation was found between abduction and lateral rotation ( P< 0.05). The greater the position of external rotation, the more abduction achieved. In conclusion, the traditional emphasis on good turnout has some scientific merit and functional implications.
The ability to externally rotate the hip is fundamental to all forms of dance, especially ballet. In ballet, steps and movements proceed and terminate from five basic positions of the feet that are largely determined by external hip rotation or turnout (Fig. 1).6, 12 In the first position, for example, the hips are usually forced to their limits of external rotation.&dquo; Every classical dancer aims to achieve perfect turnout, which is defined as &dquo;the ability of a dancer to turn his or her feet and legs out from the hip joint to a 90° position. This turnout is one of the essential principles of classical dance, giving the dancer freedom of movement in every direction.&dquo;5 Together with the pelvis, the hip is the focal point of the dancer’s concentration. It is perhaps the most
purpose? Previous publications have described problems associated with insufficient hip motion, sometimes compounded by a late start in ballet with subsequent inability to alter hip anteversion or ligamentous structures.2, 3, 13,15 However, we were unable to find any studies done to determine the ideal functional hip range for ballet. Specifically, there is no information on the point at which additional external hip rotation effectively increases the amount of obtainable hip abduction. We investigated this relationship in this study, hypothesizing that the traditional emphasis on extreme
t Address correspondence and repnnt requests to’ Shirley Kushner, Glen Sather Umversity of Alberta Sports Medicine Clinic, E-05 Van Vliet Centre, Edmonton, Alberta, T6G 2H2 Canada. 286
287
turnout may not be necessary to obtain maximum abduction of the hip.
METHODS As part of preseason medical screening, all members of the Alberta Ballet Company were studied. Twenty-two professional dancers (8 males, 14 females) from the company had bilateral hip abduction measured with increasing amounts of external rotation. After testing procedures were explained to the dancers, they performed an informal, traditional warmup. Testing began when the dancer felt that maximal flexibility had been obtained. Goniometric measurements were taken with the dancer in the supine position. Pelvic stabilization was achieved by hanging the contralateral leg over the table while the anterior superior iliac spine was held down (Fig. 3). Abduction was measured with a long arm goniometer (Goniometer, Fred Sammons Inc., Brookfield, IL), the axis of which was placed over the center of the hip joint. The center was determined by palpating the femoral artery immediately distal to the inguinal ligament. The fixed arm was aligned parallel with the body midline, while the moving arm was aligned with the long axis of the thigh between the centers of the hip and knee joints. Rotation was controlled by a second investigator with a Leighton flexometer (Leighton Flexometer, Spokane, WA) strapped to the sole of the dancer’s foot, which was held at 0° of dorsiflexion. The Leighton flexometer is a goniometer comprised of a weighted 360° scale and a weighted pointer that move independently of each other. This device was used because of its high degree of reliability.7, 10 The axis of motion was centered over the second metatarsal head so that in neutral hip rotation the foot was perpendicular to the floor
Figure 1. Five basic foot positions in ballet.
was locked at zero, with the pointer indicating 0°. The second investigator then maintained the selected external rotation position by holding the calcaneus and maintaining the subtalar joint in a neutral position while the knee was in extension. The test positions were in sequence : 0°, 45°, 60°, 70°, 80°, 90°, and maximal rotation. In each of these positions the hip was fully passively abducted. Rotation was rechecked and the amount of abduction recorded with the long goniometer arm for each of the seven test positions. Hip flexion was prevented until the position of maximal external rotation was reached. At this point, approximately 20° of flexion was allowed, since without this flexion, the dancers found it impossible to achieve maximal abduction. Both hips were tested, but right and left hips were measured in a random order.
and the scale
RESULTS The mean age of the dancers was 19.4 years for females and 24 for males. They had 12.7 and 8.5 years of training,
respectively (Table 1). None of the dancers had a history of hip pain severe enough to require either medical treatment or a day off dancing in the 6 months before testing. Analysis was done with t-tests and Pearson correlation coefficients using a 0.05 level of significance. The t-tests were performed in order to determine whether the male and female dancers differed significantly. There was a significant difference between males and females for age, height, weight,
and years in ballet (Table 1). Dancers were classified according to dominant and nondominant sides as determined by right and left handedness. All but two of the females and one of the males were right-handed. There was no significant difference between the sexes for amount of abduction achieved with the dominant or nondominant leg at 0°, 45°, 60°, 70°, or 80° of external rotation (Table 2). There was a significant difference (P < 0.05) between the sexes at 90° of external rotation for abduction achieved with the dominant limb, with greater abduction shown by the female group. There was a tendency toward significance in the nondominant limb. Pearson correlation coefficients were used to determine association between lateral rotation and abduction. A regression line was fitted showing the abduction values obtained at each level of rotation. This line of best fit was linear and there was a significant positive correlation between abduction and lateral rotation (P < 0.05). As the lateral rotation increased, degrees of abduction also increased (Fig. 4). Both males and females were able to attain more abduction in their nondominant limbs than in the dominant ones. There was a significant difference (P < 0.05) in abduction values between rotation at 0° with each of the rotation values at 45°, 60°, 70°, 80° and 90° as analyzed by paired t-tests.
288
Figure 2. Position of the pelvis and the head of the femur (Adapted from Sparger.19)
in the acetabulum
DISCUSSION Classical ballet emphasizes the importance of turnout for both aesthetic reasons and functional considerations. This study verifies the positive correlation between external rotation and the ability to achieve a greater range of abduction. The empiric wisdom of dance masters in encouraging external rotation is rewarded by facilitating abduction. Increased abduction values were present in the nondominant limbs of both male and female dancers (Fig. 4). This may be due to tighter and possibly stronger extremities on the dominant sides. The amount of abduction in this study was greater in females than in the male dancers for all of the external rotation values (Table 2). The remarkable flexibility
during (A) early
abduction and
(B)
late abduction.
achieved by dancers is linked in part to the age at which dance training is initiated. In this study of professional dancers, the females began their training on average at 6.6 years and the men at the age of 15.4 years. These ages correspond to data recorded in other series. 13 The fact that males started serious dancing during adolescence is probably one of the factors contributing toward their inability to achieve the same external rotation and hip abduction values as the females.&dquo; Miller et al. 13 have reported that male ballet dancers suffer more knee pain than females. This may be due in part to lack of hip external rotation and subsequent greater stresses on the medial joint line of the knee. Another contributing factor might be the more frequent and larger jumps, more complicated smaller jumps, and the superimposed stresses
289
Figure 3. Test position with the contralateral limb hanging over the table edge to assist in pelvic stabilization in (A) early abduction, (B) late abduction, and (C) following the 20° of flexion allowed. TABLE 1
Physical characteristics of male and female
dancers
TABLE 2
Mean hip abduction at various degrees of external rotation
Figure 4. As the external rotation increases, abduction.
so
does the
290
hip contribution, accounting for 70%
of turnout in beginners
and 68% in advanced dancers. These authors concluded that since the ideal alignment is not totally achievable at the hip, the turnout may result in significant stresses at the knee, ankle, and foot. There may be some relationship between the increasing contribution to turnout by the knee, foot, and ankle and increasing injury in these areas.11, 15, 16 In this study, although we attempted to limit the turnout of the hip by having the subjects maintain full knee extension and control foot position, some movement probably occurred at
5. The main determinants of turnout. Hardaker et al .6)
Figure
(Adapted from
TABLE 3
Range of abduction available with maximum external rotation (in degrees)
of lifting in the male dance role. The ligamentous structures of younger individuals are more adaptable. Sammarcol7 has reported that children who start dancing between 6 and 12 years of age acquire their turnout during the period when femoral neck-shaft relationships are maturing and that, after 11 years, the shape of the femoral neck cannot be altered. Retroversion, thus, cannot be increased.&dquo; Dancers who start after 11 years of age probably increase flexibility by stretching the hip capsule.2 Miller et al.l3 have reported that in the adult dancer, extraordinary amounts of external rotation have not been based on femorai neck retroversion when tested radiologically. If one concedes that hip range is acquired by task-related activities, it is interesting to review the factors that potentially limit or predispose to hip motion (Fig. 5). The extent of the turnout is limited by: 1) the femoral anteversion angle-the greater the anteversion, the less the range achievable ; 2) capsular flexibility of the hip, which, in turn, is related to generalized ligament laxity and collagen type, 3) the orientation of the acetabulum-the more anterior the inclination, the earlier the impingement of the femoral neck on the posterior lip of the acetabulum, and 4) the flexibility of the muscles and tendons crossing the hip joint.6, 12 Obviously, some of these factors are subject to training influence, whereas others are fixed. While 60° to 70° of turnout occur at the hip, some 20° to 30° may be accommodated by knee and foot alignment.’ Meinel and Atwaterll have reported that 55% of beginners and 52% of advanced or experienced dancers’ straight-legged turnout occurs at the hip. In the demi-plie, there is a greater
these joints. The dancers in our study group (males and females considered respectively) were homogenous with very little variation in height, weight, age, or hip abduction at each of the rotation values. This may support claims that dancers’ flexibility is a reflection of candidate selection in which individuals who are more flexible may become elite dancers.’ Hardaker et al.6 describe a &dquo;Darwinism of dance&dquo; whereby the professional dancer, like the elite athlete, represents the successful apex of a large field of dance students, most of whom have failed to pursue their careers because of anatomical variations such as decreased flexibility, body type, femoral anteversion, genu valgum or varum, leg length discrepancy, rigid pes planus, and pes cavus. Many dancers reported episodes of a locking or catching sensation in their hips during abduction of the free limb while dancing. Although we did not attempt to examine turnout
during weightbearing, impingement type pain
was
recorded at the limit of the range of passive abduction. During testing, three of the females and one male reported acute sharp pain during passive motion when at the limit of passive abduction in 70° to 90° of rotation. This pain was so severe that they immediately internally rotated their hips, freeing their legs from the second examiner’s hand. The pain was instantly resolved and the dancers described this mechanism and pain similar to that actively acquired when abduction was performed during dance. The pain may have been caused hv the ilinn-nnq tpnrlnn slipping nvPr an n«eous ridge on the lesser trochanter or the pectineal eminence or from the iliofemoral ligament snapping over the femoral head.8, 18 At the limits of external rotation and abduction, some dancers wanted to assume the slightly flexed position; this had to be actively controlled by the examiners. In dance, abduction movements would be done in slight hip flexion rather than in a purely frontal plane. Therefore, up to 20° of flexion was permitted so that maximal abduction could be obtained in a more realistic dance position (Fig. 3, B and
C). In situations where dancers apparently could abduct al180°, we believe that the movement was actually achieved by a combination of abduction and external rotation followed by flexion with internal rotation to return the limb to neutral (Fig. 3C). This motion is similar in many ways to Codman’s paradox at the shoulder, where rotation, abduction, and flexion maneuvers are coupled as consequential movements that bring about an involuntary movement of spin about the longitudinal axis of the humerus.I most
291
The sequence of events at the hip includes abduction with increasing external rotation to avoid contact of the greater trochanter against the ilium. At maximum abduction, there is tightness of the capsular structures and possibly impingement of the neck on the posterior acetabular lip. Further abduction at the hip can then only be achieved by slight flexion, which removes the contact of the neck because it is linked by consequential motion to internal rotation. This simultaneously relaxes the anterior capsule, permitting the completion of full abduction in slight flexion. Only the extraordinary length of the hamstrings in dancers permits the enormous arc of motion. This explains the apparent neutral position of the foot in figure 3C. The reported maximal rotation values are high and are probably a sum of full external rotation and the internal rotation that occurred as a result of the consequential movement (Table 3). These opposite rotations were probably recorded on the Leighton flexometer as total rotation. This explains why such high external rotation values were recorded (Table 3). These may be erroneous values. We recommend that in future studies the point at which the dancer needs to flex to continue abduction be the point at which maximal external rotation be recorded. We are not reporting on the significance of this data; however, it is interesting to note that a positive correlation was found between abduction achieved and maximal external rotation. A positive correlation was also found between years in ballet and the maximum rotation achieved. We have previously reported that internal rotation and hip adduction is frequently less in dancers than in the normal athletic population. 16 Furthermore, the older and more experienced the dancer, the more this trend is exaggerated and the greater the range of passive hip external rotation possible. 12, 16 We believe that the influence of external rotation on abduction ceases at the point at which flexion has to be allowed to achieve further elevation. For this reason, abduction in a functional dance situation is supplemented by early side flexion of the pelvis and flexion of the hip (Fig. 2). In summary, when testing passive hip elevation through abduction in ballet dancers, the greater the position of external rotation or turnout, the more abduction is achieved. Therefore, we conclude that the traditional emphasis on a
good turnout
has
some
scientific merit and functional im-
plication. ACKNOWLEDGMENTS We would like to thank Sharon Krantz, Ballet Mistress, and Bert Chesworth for their advice and assistance and the Alberta Ballet Company of Edmonton, Alberta for their
cooperation. REFERENCES Basmajian JV, MacConaill MA: Muscles and Movements. A Basis for Human Kinesiology Baltimore, Williams & Wilkins, 1969, p 38 2. Bergfeld JA (Moderator): Medical problems in ballet A round table. Physician Sportsmed 10(3). 98-114, 1982 3. Chmelar RD, Schultz BB, Ruhling RO, et al A physiologic profile comparing levels and styles of female dancers. Physician Sportsmed 16(7) 87-96, 1.
1988 4. Dolinar L. Ballet? MD finds it
rough
as
any sport. Physician Sportsmed 4:
86-92,1976 5 Grant G. Technical Manual and Dictionary of Classical Ballet. Third edition. New York, Dover Publications, 1982, p 122 6 Hardaker WT, Enckson L, Myers M: The dancer as athlete, in Shell C (ed) 1984 Olympic Scientific Congress Proceedings, Vol 8, Champaign, II, Human Kinetics, 1986 7 Hsieh CY, Walker JM, Gillis K: Straight-leg-raising test. Comparison of three instruments. Phys Ther 63: 1429-1433, 1983 8. Jacobs M, Young R: Snapping hip phenomenon among dancers Am Corr Ther J 32 92-98, 1978 9. Lawson J: Teaching Young Dancers—Muscular Coordination in Classical . London, Adam & Charles Black Publishers Ltd, 1984 Ballet 10. Leighton JR: An instrument and technic for the measurement of range of joint motion. Arch Phys Med Rehabil 36. 571-578, 1955 11. Meinel KK, Atwater AE. Analysis of components of the "turnout" in beginning and advanced ballet dancers. Med Sci Sports Exerc 20 (Suppl) 2: S-
6, 1988 12
Micheli LJ, Gillespie WJ, Walaszek A. Physiologic profiles of female professional ballerinas Clin Sports Med 3. 199-208, 1984 13 Miller EH, Schneider HJ, Bronson JL, et al: A new consideration in athletic 181-191, 1975 injuries. The classical ballet dancer Clin Orthop 111: 14. Peterson D: Learning through observation: The dancer’s technique, in Peterson D, Lapenskie G, Taylor A (eds): The Medical Aspects of Dance London, Ontario, Sports Dynamics, 1986, pp 49-62 15 Reid DC: Prevention of hip and knee injuries in ballet dancers. Sports Med 6 295-305, 1988 16. Reid DC, Burnham RS, Saboe LA, et al: Lower extremity flexibility patterns in classical ballet dancers and their correlation to lateral hip and knee injuries Am J Sports Med 15. 347-352, 1987 17. Sammarco J: The dancer’s hip Clin Sports Med 2 485-498, 1983 18 Schaberg JE, Harper MC, Allen WC The snapping hip syndrome. Am J Sports Med 12 361-365, 1984 19 Sparger C: Anatomy and Ballet. London, Adam and Charles Black Publish-
ers Ltd, 1975, pp 24-25 20. Vincent L The Dancers Book of Health. Kansas City, Sheed Andrews & McMeel Inc, 1978, pp 109-112 21. Wheeler L Common musculoskeletal dance injuries. Chiropractic Sports Med 1. 17-23, 1987
