Reliability and validity of goniometric turnout measurements compared with MRI and retro-reflective markers.
There is no consensus on a valid and reliable method of measuring turnout. However, there is a building awareness that such measures need to exist. Total turnout is the sum of hip rotation, tibial torsion, and contributions from the foot. To our knowledge, there has been no research that directly measures and then sums each individual component of turnout to verify a total turnout value. Furthermore, the tibial torsion component has not previously been confirmed by an imaging study. The purpose of this study was to test the validity and reliability of a single total passive turnout (TPT) test taken with a goniometer by comparing it with the sum of the individual components. Fourteen female dancers were recruited as participants. Measurements of the subjects' right and left legs were gathered for the components of turnout. Tibial torsion was measured using Magnetic Resonance Imaging (MRI). Retro-reflective marker assisted measurements were used to calculate the static components of TPT. Hip external rotation, TPT, and total active turnout (TAT) were measured by goniometer. Additional standing turnout values were collected on rotational disks. Tibial torsion and hip rotation were summed and compared with three whole-leg turnout values using Two-Tailed T-Tests and Pearson productmoment correlation coefficients. Tibial torsion measurements in dancers were found to demonstrate substantial variation between subjects and between legs in the same subject. The range on the right leg was 16[degrees] to 60[degrees], and the range on the left leg was 16[degrees] to 52[degrees]. Retro-reflective markers and biomechanical theory demonstrated that when the knee is extended and locked, "screwed home," it will not factor into a whole-leg turnout value. TAT and turnout on the disks were not statistically significant when compared with the summed total. Statistical significance was achieved in four of the eight measurement series comparing TPT with the summed value of tibial torsion and hip rotation. The advantages of a standard, valid, and reliable method of measuring turnout are many, and the risks are few. Some advantages include improved training techniques, mastery of the use of turnout at an earlier age, better dancer and teacher compliance with suggested turnout rates, understanding the use of parallel position, understanding the etiology of many dance-related injuries, and possible development of preventative measures.
There is no consensus within the dance medicine and science community on a valid and reliable method of measuring turnout. However, there is a building awareness that such measures need to exist. (1-5) Total turnout, or whole-leg turnout, is thought to be the sum of hip external rotation, tibio-femoral (knee joint) rotation, tibial torsion (twisting of the tibial shaft along its long axis), and out-toeing concurrent with dorsiflexion. (1,6) Some authors have stated that up to 40% of turnout is generated by contributions below the knee. (7) To our knowledge there has been no research that directly measures and then sums each individual component of turnout to verify a total turnout value. Furthermore, the tibial torsion component has not previously been confirmed by an imaging study. This is critical because clinical methods of measuring tibial torsion are thought to be less valid and reliable. (8-11) The dynamic components of bone (slight movement within the bone) have not been reported as a factor in turnout summation, nor has a method to measure them in static and dynamic dance positions been suggested or validated. There is emerging evidence that active and passive turnout range of motion (ROM) may be discrepant. (1,2,4,12) However, there is no gold standard for valid and reliable passive and active turnout measures because of the above mentioned uncertainties and the overall complexity of studying a motion that involves the entire lower extremity from the hip joint through the forefoot.
The purpose of this study was to test the validity and reliability of a total passive turnout (TPT) test taken with a goniometer. Lacking an absolute gold standard for turnout measurement, the individual components of turnout were gathered, summed, and then compared with TPT, total active turnout standing (TAT), and turnout on rotational disks. We hypothesized that the sum of the individual components would yield the same turnout value as when the leg was measured as one fixed unit. The secondary purpose of the study was to determine if there are correlations between excessive tibial torsion values and years of dance before ossification, handedness, turning preference, and age at first menarche.
A multi-specialty team was created that included members in musculoskeletal radiology, biomechanics, sports medicine, physical therapy, and statistics. College Institutional Review Board (IRB) approval was granted. Hospital IRB approval was granted based on the College Institutional Review Board's recommendation. Fourteen female college dancers (28 legs), who volunteered in response to group email and class announcements, provided informed consent. Criteria for inclusion were female gender, participation in ballet class at least two to three days a week by age 10, and injury free at the time of the study.
Measurements were gathered for the components of turnout and other lower extremity motions. Tibial torsion was measured using Magnetic Resonance Imaging (MRI). Measurements with retro-reflective markers calculated the static components of TPT. Hip external rotation, TPT, and TAT were measured in the clinical setting by goniometer. Additional standing turnout values were collected on rotational disks. We examined the significance of tibial torsion through a basic history that was collected by questionnaire during the initial consent phase. This included a thorough dance history documenting the amount and genre of dance before high school (14 years of age), right or left handedness, turning preference, and age at first menarche. One experienced physical therapist and a second-year physical therapy student obtained two goniometric measurements for each of the above mentioned measurements and reported the average. The averages of hip rotation measured prone and tibial torsion were summed. The averages of hip rotation measured seated and tibial torsion were also summed. These values were compared against the average TPT values, TAT values, and turnout on rotational disks using Two-Tailed T-Tests and Pearson product-moment correlation coefficients.
The testers were blinded to MRI results, and made no attempt to incorporate MRI results or remember the goniometric values as additional testing was being conducted.
Magnetic Resonance Images
T1 axial MRI images were obtained serially through the knee and ankle of each dancer's right and left tibia (28 tibiae) using a CT analogous MRI technique. (8,10) Each dancer's MRI was read by a board certified musculoskeletal radiologist.
The proximal measurement was taken one slice superior to the proximal tibiofibular joint, as described by Scheinder and Tamari. (8,10) A line was drawn parallel to the posterior cortex of the posterior tibia, and the proximal angle was measured between this posterior cortical line and the reference horizontal. The distal angle was taken at a slice best matched to the distal tibial plafond. A line was drawn from the mid-point of the anteroposterior dimension of the medial malleolus through the mid-point of the anteroposterior dimension of the distal tibiofibular joint. The distal angle measurement was the angle measured between the transecting line and the reference horizontal. This technique was slightly different than the method described by Schiender, but felt to be more repeatable. (8,10) The tibial torsion value is the proximal angle subtracted from the distal angle. The tibial torsion values were calculated using mean and standard deviation (Fig. 1).
During turnout, movement was measured at five selected locations using retro-reflective markers and photographic images: the greater trochanter of the femur, the adductor tubercle of the femur, the tibial tuberosity, the medial malleolus of the tibia, and the second toe. The main assumption was that each of these five points was moving in the transverse plane. There may also be some movement in the sagittal plane, due to the manual nature of the measurement technique and knee hyper-extension, that was neglected in this study.
Five pointers were prepared, each with a length of 60 cm. Five colored balls were placed on each pointer, and their exact locations were recorded. The tip of each pointer was touching one of the predefined locations, where it was held by an assistant. When turnout was applied by the tester, photographic images were acquired at several angles: 10[degrees], 30[degrees], 50[degrees], 60[degrees] and 70[degrees]. Each image has 2048 x 1536 pixels; thus, spatial resolution was about 0.3 mm per pixel. The acquired images were processed by a specially developed MATLAB routine that measured the location of each ball on each pointer, and calculated the location of the pointer's tip (Fig. 2).
[FIGURE 1 OMITTED]
Measurements were taken of hip external rotation prone with the knee flexed to 90[degrees], and hip external rotation seated with the hip and knee flexed to 90[degrees]. These were taken by goniometer using standard techniques. (13,14) Measurements were also taken of the dancer's first position total turnout (hips externally rotated, knees extended, heels together, and toes apart), using three different approaches: passive (TPT), active (TAT), and turnout on rotational disks. (1)
The TPT was measured by goniometer with the dancer supine on a table to approximate standing first position (1) (Fig. 3). This position allowed the spinal and pelvic musculature to remain relaxed, with pelvic rotation blocked by the table and the subject's hands. The single active component of the test, quadriceps recruitment, was used to lock the knee joint. The ankle was locked in dorsiflexion and the sub-talar joint in neutral by the tester grasping the medial and lateral malleolus with one hand and symmetrically gliding the talar dome posteriorly though the ankle mortise with the ipsilateral hand. The testers used the following technique: the thumb grasped the medial calcaneus and the tuberosity of the navicular, while the fingers simultaneously controlled the lateral calcaneus to maintain a locked hindfoot and obstruct pronation and forefoot abduction. This was considered to be necessary because it is common for dancers to dorsiflex and simultaneously abduct the forefoot, giving the illusion of additional turnout. With the hindfoot and knee locked, and the pelvis blocked in neutral, the lower extremity functioned as a fixed unit. The calcaneus was used as a lever to rotate the dancer's lower extremity externally until a capsular end feel was noted in the hip joint. (1) Each dancer was given these verbal instructions: "Maintain a level pelvis by placing the hands on your hips and keep the pelvis flat to the ceiling. Lock your knee enough that it doesn't get twisted, but relax the hip." Additional verbal cues were provided if the dancers attempted to rotate the pelvis, actively turn out, or dorsiflex and abduct the forefoot. Dancers complied easily and painlessly with these instructions. The testers visually monitored the position. The goniometer was placed with the stationary arm along the sagittal plane and the moving arm along the second metatarsal. Turnout values were noted.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Total active turnout standing (TAT) was measured by goniometer (1) (Fig. 4). The dancers were instructed to stand in first position at the barre as they do in dance class. The dancers faced the barre, placing both hands on it. No additional cues were offered. As with the passive turnout tests, active turnout measurements were taken by placing the stationary arm of the goniometer along the sagittal plane and the moving arm immediately above the second metatarsal. Turnout values were noted.
[FIGURE 5 OMITTED]
Turnout was measured in first position on the rotational disks1 (Fig. 5). The dancers were given instructions for standing on the disks to insure that each foot was similarly placed. The second metatarsal was in line with the center of the foot plate, and the medial and lateral malleoli were placed along the axis of the foot plate. Corrections and verbal cues for neutral pelvic and foot alignment were offered by the testers if the dancers were unfamiliar with the disks. Turnout values were noted.
Using a unique MRI data base, tibial torsion measurements in dancers were found to demonstrate substantial variation between subjects and between legs in the same subject (Table 1). The range on the right leg was 16[degrees] to 60[degrees], and the range on the left leg was 16[degrees] to 52[degrees]. Right tibial torsion was, on average, 2[degrees] greater than left tibial torsion (Table 2). One dancer, Subject 6, had symmetrical tibial torsion. More than five of the 14 dancers had right-left differences greater than 8[degrees]. A 17[degrees] difference between the right and left tibia was noted in Subject 10, and a 16[degrees] difference in Subject 2. Only one of the dancers with tibial torsion greater than 35[degrees] or asymmetrical tibial torsion had a history of patellofemoral complaints. The others had no history of knee pain. Three dancers had hip rotation values equal to tibial torsion values, and one dancer had less hip rotation than tibial torsion.
Knee Motion and Retro-Reflective Markers
After the knee was extended and locked (known as the "screw home mechanism"), the adductor tubercle of the femur and the tibial tuberosity moved together as the lower extremity was externally rotated. The findings demonstrate that the distal femur and the proximal tibia (the knee joint) move as a single unit when the knee is extended. Therefore, due to the screw home mechanism, the knee does not factor into a whole-leg turnout test, as the tibia will not externally rotate on the femur beyond neutral.
Range of Hip Measurements
Hip external rotation values were similar to those found by other researchers, such as Hamilton. (7) The prone range was 34[degrees] to 58[degrees] on the right and 36[degrees] to 58[degrees] on the left. The seated range on the right was 37[degrees] to 58[degrees], and 35[degrees] to 56[degrees] on the left (Table 3).
Summed Total Turnout Value = Hip External Rotation + Tibial Torsion
Four prone hip external rotation values were obtained for the right hip. These values were averaged and then summed with right tibial torsion for a total turnout value. This procedure was repeated on the left leg, and then repeated again to obtain summed total turnout values with hip external rotation measured seated. The range of summed total turnout with hip rotation measured prone yielded values from 69[degrees] to 100.75[degrees] right and 70.5[degrees] to 99[degrees] left (Table 4). The range of summed total turnout with hip rotation measured seated was 71[degrees] to 108[degrees] right and 74[degrees] to 102[degrees] left. There is no gold standard for a clinical measurement of out-toeing concurrent with dorsiflexion in dancers, so foot values were not factored into the summed total. The foot was maintained in sub-talar neutral throughout TPT testing to prevent additional degrees of turnout resulting from forefoot abduction.
Comparison of Turnout Measurements
1. Hip external rotation summed with tibial torsion (summed total turnout) compared with turnout standing in first position (TAT) When the averaged values of hip rotation measured prone were summed with tibial torsion (right summed total turnout and left summed total turnout) and compared with turnout standing in first position on the right leg and then on the left leg, no statistical correlation was found (Table 5). The same was true when the averaged values of hip rotation measured seated were summed with tibial torsion and compared with standing turnout on the right leg and the left leg. When standing turned out in first position, 75% of the dancers were either over or under turned out by 8[degrees] to 32[degrees] compared with their summed total turnout value. The remaining 25% were within 0[degrees] to 8[degrees]. However, no uniformity was noted in the results, as some subjects were standing over and others under their summed total turnout value.
2. Hip external rotation summed with tibial torsion (summed total turnout) compared with turnout on rotational disks
On the rotational disks we found 100% of the dancers' turnout values were under the measured and summed total turnout value. The dancers averaged 27[degrees] under turned out. The average under turned out on the right was 24[degrees] and on the left was 30[degrees] (Table 5).
3. Hip rotation summed with tibial torsion (summed total turnout) compared with total passive turnout (TPT)
When the summed measurements (averaged hip rotation values summed with tibial torsion values) are compared with the total passive turnout measured supine values, more correlation is found for both the right and left legs (Table 6). All the dancers' TPT values are either equal to or under the summed value. No TPT values are over the summed value. Fifty percent of the TPT values are between 0[degrees] to 5[degrees] under, and 75% are 0[degrees] to 8[degrees] under the summed total. The remaining 25% (3 dancers) demonstrated TPT values greater than 8[degrees] under the summed total turnout value. Two of these dancers were injured and unable to completely lock their knees during testing. Further, the testers could not apply adequate force to the calcaneus to externally rotate the leg without subjecting the dancer to risk of re-injury. These knee injuries were believed to have affected the results. The injured dancers were retested after six weeks and improved in some categories; however, the testers were not blinded to the MRI and other values for the second round of testing. Therefore, the improved results are discussed as an interesting artifact, but not reported in these statistical findings (Tables 7 and 8). Finally, one dancer's summed total was 20[degrees] greater than the TPT value. Though included in the overall statistics, that finding was unusual.
4. Comparison TPT, TAT, and turnout on Rotational Disks
Correlations for three total turnout measures (TPT, TAT, and turnout on rotational disks) yielded the following results. TAT was statistically significant when compared with turnout on disks (tester #1, p < .01, and tester # 2 p < .05). Turnout measured on the rotational disks was uniformly under turned out, averaging 27[degrees] compared to standing active turnout (TAT). TAT measurement was over or under the TPT; therefore, no correlation was found.
Tibial Torsion Associations
There was no correlation found between years of dance and tibial torsion: right tibial torsion with years of dancing (r = .26, p = .37) and left tibial torsion with years of dancing (r = .12, p = .70). Age of menarche, specifically delayed menarche, was not found to correlate with tibial torsion: right tibial torsion with age of menarche (r = .17, p = .64) and left tibial torsion with age of menarche (r = -.39, p = .26). One hundred percent of the subjects exhibited right hand dominance.
Reliability was addressed between testers. The student physical therapist received two 15-minute instructional sessions practicing the TPT measurement technique. Her reliability was excellent with the right (r = .78, p < .01) and left (r = .77, p < .01) for measurement comparisons with hip rotation measured prone. The untrained student tester received less than five minutes of instruction on the seated hip rotation measurement technique, and her values differed markedly.
Standardized measurement of active and passive turnout has yet to be established and, as previously noted, there is no general consensus on a valid and reliable method of measuring turnout. (1,4-6) This study attempted, whenever possible, to use those measurement techniques that have gained some currency as the gold standard. These include hip rotation measured prone and seated, and CT analogous MRI techniques for evaluating tibial torsion. (8-11,13,14) This research went further, however, to investigate the components of turnout with the knee extended. Thus it ran head-on into the problem that there is no standard clinical method to measure the "screw home" mechanism as the knee moves into full extension. To arrive at a solution biomechanical theory and retroreflective markers were used, which determined that the tibia externally rotates and screws home on the femur from a position of relative internal rotation in flexion, ending in neutral once the knee is fully extended, beyond which it will not rotate. (13,15-19) From this we conclude that the screw home mechanism will not factor into a whole leg turnout value. The testers manually controlled the subtalar joint to assure it was in neutral (not inverting or everting) and the forefoot was not abducting during passive testing. Therefore the sum of hip rotation and tibial torsion should be very close to the total passive turnout value, with no more than 0[degrees] to 5[degrees] contributed from out-toeing or in-toeing with the ankle and hind foot locked in dorsiflexion and neutral, respectively. (20)
There are some considerations for devising and utilizing information gathered from an active and passive system for measuring turnout that involves the entire lower extremity. The results reported above give rise to discussion of the following potentially important aspects of turnout measurement: the impossibility of true passive turnout testing; asymmetries or variation in turnout range of motion; the implications of excessive tibial torsion; passive and active turnout discrepancies; the arbitrary nature of 90[degrees] of turnout per leg as the limit or ideal; and the dynamic properties of bone. With these issues in mind, our findings suggest conclusions that seem contrary to traditional notions regarding the care and training of dancers. These will also be discussed.
True Passive Turnout Testing
The authors are aware that absolutely true passive turnout measurement is impossible. True passive motion means that movement is produced by the tester without assistance from the subject. (13,14) To measure passive turnout the quadriceps muscle group must be recruited to engage the screw home mechanism and lock the knee joint. Otherwise the tibia will rotate externally on the femur until the soft tissue slack is taken up. This rotation is in excess of that available when the screw home mechanism is engaged, and will generate more turnout than is biomechanically possible when standing. (18,19) External rotational forces applied to an unlocked knee joint may be unsafe for the joint structures. Fortunately, near passive turnout can be tested. Due to a high degree of coordination, dancers are capable of actively locking the knee joint, or "screwing the knee home," while simultaneously relaxing the musculature associated with hip joint rotation. Then the dancer's lower extremity can be passively turned out as one fixed unit by using the calcaneus as a lever and the hip joint as the axis of rotation. This maneuver, TPT, best approximates a whole-leg turnout motion. (1) It was therefore used in this study and checked for validity and reliability.
Active turnout is easier to measure. The dancer simply rotates the lower extremity externally using muscular effort. In our study the testers monitored for compensatory malalignments that may alter turnout values, such as anterior pelvic tilt, an unlocked knee, or pronation (including forefoot abduction).
Asymmetries or Variation in Turnout Range of Motion
There may be turnout value differences between the right and left leg due to varied amounts of warm up, muscle imbalances in strength or tightness, pelvic malalignments, the shape of the bones associated with the hip joint, or tibial torsion. (1,3,8,21-25) Some of these issues are correctable. It is possible that turnout should be measured when the hip joint is warmed up because of increased capsular laxity; also, it simulates dancing. There may be right-left imbalances between the internal and external rotator muscle groups' strength or length. Both pelvic malalignments and sacroiliac dysfunction are common in dancers, and may be correctable with medical intervention. (26) When the etiology of turnout asymmetry relates to the shape of bone, it probably cannot be altered. (23,24) It should be noted that the anteversion angle at the hip, excessive or asymmetrical tibial torsion, can have a profound effect on a dancer's overall amount of turnout.
Tibial Torsion: Knee Over Second Toe
Excessive or asymmetrical external tibial torsion in dancers is a significant finding. Average tibial torsion in the general population ranges between 20.3[degrees] to 52.1[degrees] per leg, with the mean between 37.8[degrees] and 41.7[degrees] per leg depending on the source of information. (8,10,25) This study suggests that some percentage of the dance population has tibial torsion in the excessive range, and that right-left differences exist. Therefore, the common belief that the knee at mid-patella should always be aligned immediately superior to the second toe may need to be re-examined, not just for turned out but particularly for parallel positions. As true parallel and all positions involving plie require noticeable weight bearing chain compensations, this placement may be incorrect for some dancers (Figs. 6 through 14). (25) An obvious challenge for these dancers is to stand in true parallel with the knees extended, because it is easy to see if the foot and patella are pointing forward. (21) Another challenge is the plie position turned out and in parallel, because the knee will be medial to the second toe. The dancer may choose one of several compensatory strategies to facilitate the illusion of knee over second toe. These vary depending on what position the dancer is trying to achieve. Using true parallel with extended knees as an example, there are several compensatory options available to the dancer. First, they can internally rotate the hip joint to affect lower extremity internal rotation. This will allow the feet to point directly forward, but the patella will face inward, requiring a secondary compensation at the knee. Second, they may choose neutral hip and patellar placement, but the feet will point outward. They can then adduct the forefoot to achieve the illusion of neutral alignment. Third, some dancers may choose to slightly flex the knees to disengage the screw home mechanism, allowing tibial internal rotation to bring the forefoot more anterior. Turned out positions are generally less challenging because the knee-foot alignment is somewhat less obvious. Nonetheless, many dancers choose to compensate to attempt an approximation of ideal alignment. They may under turn out (use less hip external rotation) to minimize the appearance of forced turnout.
These compensations are biomechanically inefficient choices. Some may lead to lower extremity injury such as patellofemoral syndrome, iliotibial band syndrome, and foot and ankle problems from mal-alignment. We did not find any increase in patellofemoral complaints in dancers with excessive or asymmetrical tibial torsion in turnout-related positions. We found increased patellofemoral symptomatology evident in only one dancer with excessive external tibia torsion, and it appeared to be related to the subject's difficulty achieving a true parallel plie. When excessive or asymmetrical external tibial torsion is present, dance teachers may wish to focus on pelvic and foot placement rather than specific knee alignment.
The Dancer's Gait and Excessive External Tibial Torsion
A common belief that dancers walk with their toes pointing outward due to external and internal rotator muscle imbalances may not tell the whole story. Dancers with excessive external tibial torsion may out-toe in ambulation as do others with this bony alignment. (13)
Tibial Torsion Etiology
The etiology of the newly discovered differences in dancers' right and left tibial torsion is unclear, but if it is the result of training protocols, such as always beginning barre on the right side, corrections could easily be instituted and tracked through prospective research collected by yearly screenings. Tibial torsion rates may be possible to collect without use of MRI. If the TPT measurement is found to be valid, then hip rotation could be subtracted from TPT, leaving a difference that is largely tibial.
Passive versus Active Turnout Discrepancies
Passive and active ROM is commonly compared in physical therapy clinics to check for strength deficits. (22,27) It seems reasonable to use this same technique when investigating the strength of dancers. It is common knowledge that many dancers want to have as close to 90[degrees] turnout per leg as possible. (1-3,28,29) Many believe that less turnout will negatively effect their careers. Therefore, testers should be aware that there can be an emotional response to turnout measurement that could alter outcome. Dancers can see and feel their turnout, and therefore control how much muscular effort is used to turn the legs out. If they want additional standing turnout they simply exert increased muscular effort, screw the knees, or abduct the forefoot. Conversely, some dancers may choose to under turn out, believing it safer. For example, one dancer was asked several months after the conclusion of data gathering why she chose not to use all her available turnout. This particular dancer had significant external tibial torsion and could not get the knee over the second toe during parallel and turned out plie without significant forefoot adduction, causing peroneal muscle group discomfort (Figs. 9 through 11, 15 and 16). We were told that a childhood dance teacher who could see that the knee was not over the second toe instructed this dancer to turn out less to avoid injury. Thus, she made a conscious effort to withhold some of the available range of motion.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
There are additional reasons why a dancer may not maximally turn out. Many dancers simply lack the strength to achieve full turnout. (1,2,30-32) Several research teams have discovered active and passive ROM differences in dancers. (2,4,12,25,32) We found that when TPT is compared with the disk measurement there is a 22[degrees] difference on the right leg and a 19[degrees] difference on the left leg. (25) Welsh and colleagues found 33[degrees] hip external rotation lag of both legs combined when the whole leg measurements of TPT were compared with disk measurements. (4) Crookshanks measured hip rotation only in various age and achievement groups, and found lags up to 13[degrees] in both legs. (12) Negus found up to a 30[degrees] external rotation lag of both legs combined with hip only measurements. (2) Weakness at end range hip external rotation seems commonplace and problematic because end range turnout is precisely where dancers choose to stand. Fatigue may also play a role with successive turnout attempts when the turnout musculature demonstrates poor endurance. Coordination may have an effect on proper execution because turnout requires complex interplay between several muscle groups.
[FIGURE 12 OMITTED]
Passive measures of turnout are by definition the most objective, as the tester can manually rotate the extremity externally until all the soft tissue slack is taken up and monitor the dancer for compensations used to create the illusion of additional turnout. Further, the dancer is supine, looking at the ceiling, and therefore cannot see the degree to which the extremity is turning out. The objective passive turnout value can be compared with the active value to determine if the dancer is over or under turning out, and what compensatory strategies are being employed.
90[degrees] of Turnout
When tibial torsion values from 16[degrees] to 60[degrees] are summed with hip rotation values from 34[degrees] to 58[degrees], some of the dancers had greater than 90[degrees] turnout per leg. To our knowledge this has never before been mentioned in the dance science literature, because all whole-leg turn out measurement research has involved active tests, which are generally thought to yield lower average values than passive tests. Further, there are no prior research projects confirming tibial torsion values in dancers with imaging studies. (4) It is commonly thought that few, if any, dancers have 90[degrees] of turnout per leg. We may discover that they have more available turnout when measured passively, and lack the end range strength to control it. Dancers who know their turnout range of motion can effectively train to use it. (1,3,6)
[FIGURE 13 OMITTED]
Dynamic Properties of Long Bones
The biomechanical properties of bone may affect the absolute turnout value when stresses are applied differently. The trabecular and cortical components of long bone yield to strain at varying rates, depending on speed and amount of force. (33-35) Though this has not previously been described in the dance medicine literature regarding turnout measurement, it is worth noting that this fact may alter turnout values. A dancer standing statically may have less turnout than when different forces are applied, such as landing from a spinning leap.
Risk-Benefit Profile of Turnout Measurement
In spite of these challenges, and because turnout is a unique and integral component of many dance idioms, standard measurement techniques need to be developed. The advantages of a standard, valid and reliable method to measure turnout are many, and there are few if any risks. Some advantages might include: improved training techniques, mastering the use of turnout at an earlier age, better dancer and teacher compliance with suggested turnout rates, understanding the use of parallel position, understanding the etiology of many dance-related injuries and possible development of preventative measures. Because there are at this time no standard, valid, or reliable measurement techniques for turnout, prospective studies with large databases against which to validate these impressions are absent. It is thought by many in the dance medicine and science community, these authors included, that proper execution of turnout promotes improved dancing, while poor execution may retard dancer progress and lead to injury. For example, pronation is a common compensation of over turning out the leg, and there is a significant pool of orthopedic literature that describes the pronated foot's relationship to lower extremity and spinal injury. (36-38) Some dance medicine research also makes this connection. 39,40 With valid and reliable turnout values dance researchers may be able to draw a direct correlation between pronation and poor use of turnout, thereby basing their observations on evidence.
[FIGURE 14 OMITTED]
Some authors have discussed the challenges of transferring dance science information from the clinic or lab to the studio. (3) Such follow through might be improved if turnout values were easier to measure, and improved use could be directly linked to improved rates of skill acquisition.
There were some limitations to this study. The sample size was small. The dancers were not uniformly warmed up. One tester knew some of the subjects as students and patients. Two subjects developed knee injuries in the time period between the MRI and turnout testing. This may have affected the initial results. The subjects were retested after the injuries healed, resulting in improved correlation (Tables 6 and 7). The dancers were not blinded to the results of the active tests; they could see, hear, and feel the values as they were spoken and recorded, and this may have affected some of the results.
[FIGURE 15 OMITTED]
We hope that these tests will help stimulate progress toward a better understanding of the active and passive components of turnout, leading to a single clinical technique for measuring passive turnout that includes the hip, tibia, foot, and ankle. This measurement could improve dance training by correcting problems with over or under turning out. With valid and reliable turnout measurements dancers will not spend years frustrated by alignment problems that they cannot correct, while developing poor motor patterning in compensation. They can learn to use appropriately all the turnout they have, but not more. A valid and reliable system for measuring turnout in first through fifth positions should prevent or correct some problems associated with turnout-related injuries. Future research can address the validity and reliability of this passive technique in first position, and then modify it for all other dance positions. Further, the challenge of parallel stance and plie should be addressed by the dance community. A valid test for passive turnout could facilitate clinical tibial torsion measurement in dancers by subtracting hip external rotation from TPT. Developing evidence to validate these measurements can verify a long standing belief that better use of turnout will improve the dancer's well being.
[FIGURE 16 OMITTED]
We would like to thank: St. Luke's Hospital Department of Radiology for donating the MRI; June Marcotte, R.T. (R) (MR), for patiently providing expert imaging services; Muhlenberg College Dance Department Director Professor Karen Dearborn for her support; the intellectually curious and responsible dancers who volunteered as subjects and returned for each testing session; and our research assistant, Gabrielle Campagna. The ongoing support and expert advice from Michael Grossman, M.D., Tom Welsh, Ph.D., Steven Chatfield, Ph.D., and Marijeanne Liederbach, Ph.D., P.T, A.T.C., C.S.C.S., was invaluable.
(1.) Grossman G. Measuring dancer's active and passive turnout. J Dance Med Sci. 2003;7(2):49-55.
(2.) Negus V. Associations between turnout and lower extremity injuries in classical ballet dancers. J Orthop Phys Ther. 2005 May;35(5):30518.
(3.) Grossman G, Krasnow D, Welsh T. Effective use of turnout; biomechanical, neuromuscular, and behavioral considerations. J Dance Educ. 2005;5(1):15-27.
(4.) Welsh T, Rodriguez M, Iverson L, et al. Assessing turnout in university dancers. In: Solomon R, Solomon J (eds): Proceedings of the 17th Annual Meeting of the International Association for Dance Medicine & Science 2007. Canberra, Australia: IADMS, 2007, pp. 50-54.
(5.) Watkins A, Woodhul-Mcneal A, Clarkson P, Ebbeling C. Lower extremity alignment and injury in young, pre professional, college, and professional dancers. Med Probl Perform Art. 1989 Dec;(4)4:148-58.
(6.) Gilbert C. Relationship between hip external rotation and turnout angle for the five classical ballet positions. J Sports Phys Ther. May 1998;27(5):339-47.
(7.) Hamilton W, Hamilton L, Marshall P, Molnar M. A profile of the musculoskeletal characteristics of elite professional ballet dancers. Am J Sports Med. 1992 May-June;20(3):26773.
(8.) Tamari K, Tinley P, Briffa K, Breidahl W. Validity and reliability of existing and modified clinical methods of measuring femoral and tibiofibular torsion in healthy subjects: use of different axes may improve reliability. Clin Anat. 2005;(18)46-55.
(9.) Hudson D, Royer T, Richards J. Ultrasound measurements of torsions in the tibia and femur. J Bone Joint Surg Am. 2006 Jan;88(1):138-43.
(10.) Scheider B, Laubenberger J, Jemlich S, et al. Measurement of femoral antetorsion and tibial torsion by magnetic resonance imaging. Br J Radiol. 1997 June;70(834):575-9.
(11.) Lang L, Volpe R. Measurement of tibial torsion. J Am Podiatr Assoc. 1998;88(4):161-5.
(12.) Crookshanks D. Normative dance specific musculoskeletal parameters for young female dancers in Australia. In: Solomon R, Solomon J (eds): Proceedings of the 17th Annual Meeting of the International Association for Dance Medicine & Science 2007. Canberra, Australia: IADMS, 2007, pp. 249-52.
(13.) Magee DJ. The hip. In: Benson H (ed): Orthopedic Physical Assessment, Philadelphia: WB Saunders Company; 1987, pp. 239-265.
(14.) Norkin CC, White DJ. The hip. In: Measurement of Joint Motion: A Guide to Goniometery: Philadelphia: FA Davis Company, 1985, pp. 86-87.
(15.) Johal P, Williams A, Wragg P, et al. Tibio-femoral movement in the living knee. A study of weight-bearing and non-weight bearing knee kinematics using 'interventional MRI.' J Biomech. 2005;38:269-76.
(16.) Iwaki H, Pinskerova V, Freeman M. Tibiofemoral movement 1: the shapes and relative movements of the femur and tibia in the unloaded cadaver knee. J Bone Joint Surg Br. 2000 Nov;82(8):1189-95.
(17.) Dyrby C, Andriacchi T. Secondary motions of the knee during weight bearing and non-weight bearing activities. J Orthop Res. 2004;22:794800.
(18.) Freeman M, Pinskerova V. The movement of the normal tibio-femoral joint. J Biomech. 2005;38:197208.
(19.) Patel V, Hall K, Lotz J, et al. A three-dimensional MRI analysis of knee kinematics. J Orthp Res. 2004;22:283-92.
(20.) Richardson J, Iglarsh Z. The Foot and Ankle. In: Weinstien J, Rydivik B (eds): Clinical Orthopedic Physical Therapy. Philadelphia: WB Saunders Company, 1994, pp. 501-513.
(21.) Sahrmann SA. Movement impairment syndromes of the hip. In: White K (ed): Diagnosis and Treatment of Movement Impairment Syndromes. St, Louis, MO: Mosby, 2002, pp. 121-190.
(22.) Hoppenfeld S. Physical Examination of the Spine and Extremities. Norwalk, CT: Appleton-Century-Crofts, 1976.
(23.) Bauman P, Singson R, Hamilton W. Femoral neck anteversion in ballerinas. Clin Orthop Relat Res. 1994 May;302:57-63
(24.) Hamilton D, Aronsen P, Loken JH, et al. Dance training intensity at 1114 years is associated with femoral torsion in classical ballet dancers. Br J Sports Med. 2006; 40:299-303.
(25.) Grossman G, Waninger K, Voloshin A, et al. Reliability and validity of goniometric turnout measurements compared with MRI and retro reflective markers. In: Solomon R, Solomon J (eds): Proceedings of the 16th Annual Meeting of the International Association for Dance Medicine & Science 2006. Eugene, OR: IADMS, 2006, pp. 9-12.
(26.) Cibulka MT, Sinacore DR, Cromer GS, Delitto A. Unilateral hip rotation range of motion asymmetry in patients with sacroiliac joint regional pain. Spine. 1998;23(9):1009-15.
(27.) Andrews JM, Harrelson GL, Wilk KE. Physical Rehabilitation of the Injured Athlete. Philadelphia: Saunders, 2004.
(28.) Gilbert CB, Gross MT, Klug KB. Relationship between hip external rotation angle for the five classical ballet positions. J Orthop Sports Phys Ther. 1998 May;27(5):33947.
(29.) Clippinger-Robertson K. Biomechanical considerations in turnout. JOPERD. 1987 May/June;58(5);3740.
(30.) Kahn K, Bennell K, Ng S, et al. Can 16-18-year-old elite ballet dancers improve their hip and ankle range of motion over a 12 month period of time. Clinical J Sports Med. 2000;10:98-103.
(31.) Bennell K, Kahn K, Matthews B, Singleton C. Changes in hip and ankle range of motion and hip muscle strength in 8-11 year old novice female ballet dancers and controls: a 12 month follow up study. Br J Sports Med. 2001;35:54-9.
(32.) Gupta A, Fernihough B, Bailey G, et al. An evaluation of differences in hip external rotation strength and range of motion between female dancers and non-dancers. Br J Sports Med. 2004 Dec;38(6):778-83.
(33.) Chunjuan D, Hongshun M, Ruo M, et al. An experimental study on the biomechanical properties of the cancellous bones of distal femur. Biomed Mater Eng. 2006;16(3):215-22.
(34.) Bayraktar H, Morgan E, Niebur G, et al. Comparison of the elastic and yield properties of human femoral and cortical bone tissue. J Biomech. 2004 Jan;37(1):27-35.
(35.) Keaveny T, Guo X, Wachtel E, et al. Trabecular bone exhibits elastic behavior and yields at low strains. J Biomech.1994 Sep:27(9):1127-36.
(36.) Hardaker W. Foot and ankle injuries in classical ballet dancers. Orthop Clin North Am. 1989 Oct;20(4):621-9.
(37.) Brody D. Running injuries. Clin Symp.1980;32(4):2-9.
(38.) Marshall P. The rehabilitation of overuse foot injuries in athletes and dancers. Clin Sports Med. 1988 Jan;6(3):639-55.
(39.) Gamboa JM, Roberts LA, Maring J, Ferus A. Injury patterns in elite adolescent preprofessional ballet dancers and the use of screening data to describe and predict injury characteristics. In: Solomon R, Solomon J (eds): Proceedings of the 16th Annual Meeting of the International Association for Dance Medicine & Science 2006. Eugene, OR: IADMS, 2006, pp. 47-51.
(40.) Solomon R, Trepman E, Micheli L. Foot morphology and injury patterns in ballet and modern dancers. Kinesiol Med Dance. 1989;12(1):20-40.
(41.) Grossman G. The dancer's hip. In: Hughes C (ed): Independent Study Course 18.3, Dance Medicine: Strategies for the Prevention and Care of Injuries to Dancers. La Crosse, WI: Orthopaedic Section, APTA, Inc., 2008.
Gayanne Grossman, P.T., Ed.M., is a Physical Therapist and Associate Instructor, Department of Theater and Dance, Muhlenberg College, Allentown, Pennsylvania. Kevin N. Waninger, M.D., M.S., is the Director of the Sports Medicine Fellowship, Department of Family Medicine, St. Luke's Hospital, Bethlehem, Pennsylvania. Arkady Voloshin, Ph.D., is in the Department of Mechanical Engineering and Mechanics, Lehigh University, Bethlehem, Pennsylvania. William R. Reinus, M.D., M.B.A., F.A.C.R., is the Vice Chairman of the Department of Radiology, Planning, and Development, and Professor and Chief Musculoskeletal and Trauma Radiology, Temple University Hospital, Philadelphia, Pennsylvania. Rachael Ross, S.P.T., is at the Jefferson College of Health Professions, Philadelphia, Pennsylvania. Jill Stoltzfus, Ph.D., is the Director of the Research Institute, St. Luke's Hospital, Bethlehem, Pennsylvania. Kathleen Bibalo, is at Muhlenberg College, Allentown, Pennsylvania.
Correspondence: Gayanne Grossman, P.T., Ed.M., Department of Theatre and Dance, Muhlenberg College, 2400 Chew Street, Allentown, Pennsylvania 18014; firstname.lastname@example.org.
This research was presented at the International Association for Dance Medicine and Science 2006 annual conference.
Table 1 Tibial Torsion Raw Data Subject Right Tibial Torsion Left Tibial Torsion #1 32 31 #2 45 29 #3 39 30 #4 29 26 #5 29 24 #6 16 16 #7 31 35 #8 30 28 #9 48 45 #10 60 43 #11 44 52 #12 39 40 #13 34 37 #14 25 35 Average 36 34 Standard Deviation 11 9 Table 2 Tibial Torsion Ranges and Means N Min Max Mean Std Dev Right 14 16 60 35.79 11.033 Left 14 16 52 33.64 9.386 Table 3 Hip Rotation Ranges Min Max Right 34 58 Left 36 58 Table 4 Summed Value Hip Rotation and Tibial Torsion (N = 12) Min Max Mean Std R Prone HR + TT 69 100.75 83.50 11.63 L Prone HR + TT 70.5 99.0 81.92 8.64 Table 5 Summed Values Prone Hip Rotation and Tibial Torsion (Summed Total Turnout) compared with Standing Turnout (TAT) and Turnout on Disks Tester # 1 Tester # 2 Right Summed Turnout/TAT r = .50, p = .12 r = .53, p = .10 Left Summed Turnout/TAT r = .35, P = .29 r = .53, p = .09 Right Summed Turnout/Disk r = .16, p = .63 r = .14, p = .66 Left Summed Turnout/Disk r = .11, p = .75 r = -.22, p = .50 Combined Value Right Summed Turnout/TAT r = .56, p = .07 Left Summed Turnout/TAT r = .53, p = .09 Right Summed Turnout/Disk r = .51, p = .11 Left Summed Turnout/Disk r = .45, p = .17 Table 6 Summed Values Hip Rotation and Tibial Torsion (Summed Total Turnout) compared with TPT Tester # 1 Tester # 2 Right (hip rotation prone) r = .57, p = .07 r = .62, p = .05 Left (hip rotation prone) r = .68, p = .05 r = .31, p = .33 Right (hip rotation seated) r = .34, p = .27 r = .59, p < .05 Left (hip rotation seated) r = .28, p = .37 r = .61, p < .05 Combined Value Right (hip rotation prone) r = .60, p < .05 Left (hip rotation prone) r = .53, p = .09 Right (hip rotation seated) r = .53, p < .06 Left (hip rotation seated) r = .53, p = .08 Table 7 Post Injury Test Prone Hip Rotation and Tibial Torsion (Summed Total Turnout) compared with Standing Turnout (TAT) Tester # 1 Tester # 2 Right Summed r = .50, p = .12 initial r = .53, p = .10 initial Turnout/TAT r = .46, p = .13 retest r = .36, p = .26 retest Left Summed r = .35, p = .29 initial r = .53, p = .09 initial Turnout/TAT r = -.05, p = .88 retest r = .16, p = .63 retest Combined Value Right Summed r = .56, p = .07 initial Turnout/TAT r = .54, p = .07 retest Left Summed r = .56, p = .09 initial Turnout/TAT r = .18, p = .58 retest Table 8 Post Injury Test Hip Rotation and Tibial Torsion (Summed Total Turnout) compared with TPT Tester # 1 Right r = .57, p = .07 initial (hip rotation prone) r = .78, p < .01 retest Left r = .68, p = .05 initial (hip rotation prone) r = .61, p = .05 retest Right r = .34, p = .27 initial (hip rotation seated) r = .36, p = .27 retest Left r = .28, p = .37 initial (hip rotation seated) r = .25, p = .45 retest Tester # 2 Right r = .62, p = .05 initial (hip rotation prone) r = .80, p < .01 retest Left r = .31, p = .33 initial (hip rotation prone) retest unchanged Right r = .59, p < .05 initial (hip rotation seated) r = .74, p < .01 retest Left r = .61, p < .05 initial (hip rotation seated) r = .51, p = .09 retest Combined Value Right r = .60, p < .05 initial (hip rotation prone) r = .80, p < .01 retest Left r = .53, p = .09 initial (hip rotation prone) r = .37, p = .27 retest Right r = .56, p < .06 initial (hip rotation seated) r = .61, p = .05 retest Left r = .53, p = .08 initial (hip rotation seated) r = .39, p = .23 retest
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Original Article; magnetic resonance imaging|
|Author:||Grossman, Gayanne; Waninger, Kevin N.; Voloshin, Arkady; Reinus, William R.; Ross, Rachael; Stoltzfu|
|Publication:||Journal of Dance Medicine & Science|
|Date:||Oct 1, 2008|
|Previous Article:||Assessing turnout in university dancers.|
|Next Article:||Scand J Med Sci Sports: Operative treatment of stress fractures of the proximal second metatarsal.|