Videographic Drop-Jump Screening Test

A drop-jump videographic screening test was developed as described in detail previously.⁶⁵ Minimal equipment is required and the software is available from the authors’ nonprofit foundation. A camcorder equipped with a memory stick is placed on a stand 102.24 cm (40.25 inches) in height. The stand is positioned approximately 365.76 cm (12 ft) in front of a box 30.48 cm (12 inches) inheight and 38.1 cm (15 inches) in width. One-inch Velcro circles are placed on each of the four corners of the box that faces the camera. Athletes wear fitted, dark shorts and low-cut gym shoes. Reflective markers are placed at the greater trochanter and the lateral malleolus of both the right and the left legs, and Velcro circles are placed on the center of each patella. A research assistant demonstrates the jump-land sequence, and one practice trial is done to ensure the athlete understands the test. No verbal instruction regarding how to land or jump is provided. Athletes are told only to land straight in front of the box to be in the correct angle for the camera to record properly. The athletes perform a jump-land sequence by first jumping off the box, landing, and immediately performing a maximum vertical jump. This sequence is repeated three times.

After completion of the test, a research assistant views all three trials, and the one that best represents the athlete’s jumping ability is selected for measurement. Advancing the videographic frame-by-frame, the following images are captured as still photographs: (1) pre-land, the frame in which the athlete’s toes just touch the ground after the jump off of the box; (2) land, the frame in which the athlete is at the deepest point; and (3) take-off, the frame that demonstrates the initial forward and upward movement of the arms and the body as the athlete prepares to go into the maximum vertical jump.

The captured images are imported into a hard drive of a desktop computer and digitized on the computer screen. A calibration procedure is done by placing the cursor and clicking in the center of each Velcro marker on each of the four corners of the drop-jump box. The anatomic reference points represented by the reflective markers are selected by clicking in a designated sequence the cursor for each image.

The absolute centimeters of separation distance between the right and the left hip and normalized separation distances for the knees and ankles, standardized according to the hip separation distance, are analyzed. Normalized knee separation distance is calculated as knee separation distance divided by hip separation distance and normalized ankle separation distance is calculated as ankle separation distance divided by hip separation distance (Fig. 16-2).

FIGURE 16-2 The videographic test produced photographs of three phases of the drop-jump test. The centimeters of distance between the hips, the knees, and the ankles was calculated along with normalized knee and ankle separation distance (according to the hip separation distance). Shown is the test result of a 14-year-old female.

(From Noyes, F. R.; Barber-Westin, S. D.; Fleckenstein, C.; et al.: The drop-jump screening test: difference in lower limb control by gender and effect of neuromuscular training in female athletes. Am J Sports Med 33:197–207, 2005.)

The reliability of the drop-jump videographic test was determined in 17 female athletes who underwent the test twice, 7 weeks apart.⁶⁵ The reliability of the absolute centimeters of hip separation distances was evaluated. Hip separation distance was expected to be highly reliable, thus providing the basis for normalization of knee and ankle separation distances. Then, in 10 other subjects, reliability within the videographic test was assessed by capturing two of the three jump-land sequence trials and comparing the absolute centimeters of hip, knee, and ankle separation distances between the two sequences on the same day of testing. For the test-retest trial, the intraclass correlation coefficients (ICC) for the hip separation distance demonstrated high reliability (pre-land, .96; land, .94; take-off, .94). For the within-test trial, the ICC for the hip, knee, and ankle separation distance were all .90 or higher, demonstrating excellent reliability of the videographic test and software capturing procedures.

Isokinetic Evaluation of Knee Extensor and Flexor Strength

Isokinetic knee extensor and flexor testing was conducted (concentric mode) at 300°/sec (Fig. 16-3) in athletes over the age of 11 years. For the children who were 9 to 10 years of age, this test was performed at 180°/sec. Other investigators have reported acceptable reliability of isokinetic measurements in children and adults at these test velocities.^48,58,60 The athletes first completed a dynamic warm-up session that lasted 5 minutes. They were then positioned in the chair of the device with appropriate torso, pelvis, and thigh straps placed according to the manufacturer’s protocol. The lever arm of the dynamometer was aligned with the lateral epicondyle of the knee, with the knee flexed to 90°. The chair was adjusted for each athlete as required to ensure proper positioning. The range of motion during the test was fixed from 90° to 0°. Gravitational factors were calculated by the dynamometer and automatically compensated for during the tests. The athletes performed three to four submaximal trials to become familiar with the machine and test velocity. A total of 10 repetitions were completed and the highest peak torque value used for analyses. Mean peak torque values (in Newton-meters [Nm]) were normalized for body weight (BW) in kilograms. Verbal encouragement was given throughout the tests, because the athletes were told to kick as hard and as fast as possible, but no visual feedback was available.

FIGURE 16-3 Isokinetic knee extensor and flexor testing.

Isokinetic Evaluation of Internal and External Tibial Rotation Strength

An isokinetic assessment of knee internal tibial rotation (IR) and external tibial rotation (ER) strength was conducted on a calibrated Biodex System 3 (Biodex Medical Systems, Inc., Shirley, NY). The IR and ER peak torques of each limb were obtained at 120°/sec and 180°/sec. Testing was preceded by a dynamic warm-up that lasted approximately 5 minutes. The patients were positioned in the device in a partial supine position with the hip flexed 60° and the knee flexed 90° as described by Hester and Falkel (Fig. 16-4).³⁹ Appropriate torso, pelvis, and thigh straps were placed according to the manufacturer’s protocol. In addition, the foot and ankle were tightly secured to a footplate using Velcro (VELCRO USA, Manchester, NH) straps.³⁹ The athletes were instructed to relax so that the neutral position with regard to tibial rotation and the range of internal and external tibial motion could be obtained.^72,89 Five submaximal repetitions were done before each test to promote familiarization with the dynamometer velocity speeds and to set the range of motion limits for both IR and ER.

FIGURE 16-4 A, Subject positioning for the isokinetic testing of internal and external tibial rotation. B, Foot and ankle strapping for stabilization.

(A and B, Reprinted from Noyes, F. R.; Barber-Westin, S. D.: Isokinetic profile and differences in tibial rotation strength between male and female athletes 11 to 17 years of age. Isok Exer Sci 13:251–259, 2005; with permission from IOS Press.)

There were 2-minute rest periods between the two velocity speed tests. The testing sequence of the dominant and nondominant limbs was randomized among the subjects. Verbal encouragement was given throughout the tests, because the athletes were told to rotate their lower leg through full range with maximal effort, but no visual feedback was available. A total of eight repetitions were completed at each speed and the single highest peak torque value used for analyses. Peak torque values were normalized for BW in kilograms and were expressed as (Nm/BW). Time-to-peak torque was measured in milliseconds (msec). The ratio of the mean peak torque of IR was calculated relative to that of ER (ratio IE = IR/ER × 100) in a manner described by Segawa and associates.⁸² The ratio of the mean peak torque of both IR and ER were also calculated relative to that of knee flexion. The ratio IR-F was calculated as IR/flexion × 100, and the ratio ER-F was calculated as ER/flexion × 100. The ratio of the mean peak torque of IR (IR-E) and ER (ER-E) were also calculated relative to that of knee extension in a manner similar to that described for knee flexion.

Single-Leg Functional Hop Testing

Two single-leg functional hop tests were conducted: a timed side-hop and a crossover hop for distance. The athletes were provided with instructions and one practice trial was conducted for each test. In the timed side-hop test, a course was created on the floor using masking tape to mark three boxes that were 46 × 46 cm (Fig. 16-5).⁷ The athletes were encouraged to hop as fast as possible, but to maintain balance to be able to complete the test. If the athlete did not clear the lines, double-bounced between hops, or could not hold balance on landing, a zero was recorded. Two tests were completed for each limb, with mean times calculated with a standard stopwatch to the nearest .01 sec.

FIGURE 16-5 Single-leg timed side-hop function test. Subject begins in box A. Standing on one leg, the subject hops to box B, box C, then back to box B and box A to complete one course repetition. One trial consists of three repetitions. The subject must successfully cross the lines, maintain single-legged balance, and hold each landing without placing the opposite foot down on the floor.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

In the crossover hop for distance test, a marking strip made of masking tape was placed on the floor that extended approximately 6 m (Fig. 16-6).⁶³ The athletes hopped three consecutive times on one foot, crossing diagonally over the tape on each hop. They were encouraged to go as far as possible while maintaining balance and control. The total distance hopped was measured and each leg tested twice, with the average distance calculated. The reliability of this test has been reported by other investigators to be excellent.^5,12,29

FIGURE 16-6 Crossover hop for distance test. The athlete hops three consecutive times on one foot, crossing diagonally over the tape on each hop. Subjects are encouraged to hop as far as possible while maintaining balance and control.

(From Noyes, F. R.; Barber, S. D.; Mangine, R. E.: Abnormal lower limb symmetry determined by function hop tests after anterior cruciate ligament rupture. Am J Sports Med 19:513–518, 1991.)

Limb symmetry was calculated for each test by dividing the mean of the right limb by the mean of the left limb andmultiplying the result by 100. Prior studies on populations aged 17 to 34 reported that less than 85% limb symmetry on single-leg-hop tests was abnormal and representative of a general lower limb functional limitation.^6,63

Hypothesis #1: Female Athletes Have an Overall Lower Limb Valgus Alignment, Indicated by Abnormally Decreased Knee Separation Distances, on Landing and Acceleration into a Vertical Jump on a Videographic Drop-Jump Test

The hypothesis was supported, because a marked overall lower limb valgus alignment (indicated by a knee separation distance of ≤ 60%) was found in 63% of the female athletes on landing (Fig. 16-7) and in 80% on take-off.

FIGURE 16-7 The distribution of male athletes and the 62 female athletes before and after training according to normalized knee separation distance on landing. Whereas there was no difference in the distribution of male and untrained female athletes among the categories shown, a significant difference was present between male and trained females (P < .0001). The stick figures are representative of the knee separation distance but not of the ankle separation distance.

(From Noyes, F. R.; Barber-Westin, S. D.; Fleckenstein, C.; et al.: The drop-jump screening test: difference in lower limb control by gender and effect of neuromuscular training in female athletes. Am J Sports Med 33:197–207, 2005.)

Hypothesis #2: Male Athletes Have a More Neutrally Aligned Lower Limb Position, with a Greater Amount of Separation Distance between the Knees, on Landing and Acceleration into a Vertical Jump Compared with Females on a Drop-Jump Test

The hypothesis was not supported, as a marked overall lower limb valgus alignment was found in 75% of the male athletes on landing (see Fig. 16-7) and in 72% on take-off. There was no statistically significant difference between male and female athletes in the mean normalized knee and ankle separation distance during the landing and take-off phases. The female athletes had significantly higher mean normalized knee and ankle separation distances during the pre-land phase only.

The normalized knee separation distances for both genders are shown in FIGURE 16-8, the normalized ankle separation distances are shown in FIGURE 16-9, and the absolute values for the hip, knee, and ankle separation distances are shown in Table 16-1. There were no correlations between knee and ankle separation distances for each phase of the jump-land sequence in either the female or the male athletes.

FIGURE 16-8 The mean normalized knee separation distances for the three phases of the drop-jump test are shown for the male athletes, untrained female athletes, and trained female athletes. After training, female athletes had statistically significant increases in the mean normalized knee separation distance in all three phases (P < .001) and had statistically greater mean normalized knee separation distances than males for all phases (P < .0001).

(From Noyes, F. R.; Barber-Westin, S. D.; Fleckenstein, C.; et al.: The drop-jump screening test: difference in lower limb control by gender and effect of neuromuscular training in female athletes. Am J Sports Med 33:197–207, 2005.)

FIGURE 16-9 The mean normalized ankle separation distances for the three phases of the drop-jump test are shown for the male athletes, untrained female athletes, and trained female athletes. After training, female athletes had statistically significant increases in the mean normalized ankle separation distance in all three phases (P < .001) and had statistically greater mean normalized ankle separation distances than males for all phases (P < .0001).

(From Noyes, F. R.; Barber-Westin, S. D.; Fleckenstein, C.; et al.: The drop-jump screening test: difference in lower limb control by gender and effect of neuromuscular training in female athletes. Am J Sports Med 33:197–207, 2005.)

Hypothesis #3: A Neuromuscular-Training Program Will Produce Significantly Greater Knee Separation Distance in Female Athletes, thereby Producing a More Neutrally Aligned Lower Limb Position on a Drop-Jump Test

The neuromuscular-training program was effective, supporting the hypothesis, because statistically significant increases were found after training in the subgroup of 62 females in the absolute (see Table 16-1) and the normalized knee (see Fig. 16-8) and ankle (see Fig. 16-9) separation distances for all phases of the jump-land sequence (P < .001). After training, the female athletes had statistically greater mean normalized knee and ankle separation distances than the male athletes for all phases of the jump-land sequence (P < .0001). In addition, after training, a statistically significant difference was found in the distribution of female athletes who had less than 60%, 61% to 80%, or greater than 80% knee separation distance on landing (P = .003) and take-off (P = .006). The distribution in the percentage change in knee and ankle separation distances after training is shown for all three phases of the jump-land sequence in Table 16-2.

Another important finding of this investigation was the significant improvement after training in knee flexor peak torque values in both the dominant and the nondominant legs (P < .0001). In the dominant leg, knee flexor peak torque increased from 40 ± 8 Nm (mean ± standard deviation) before training to 44 ± 7 Nm after training; and in the nondominant leg, from 38 ± 8 Nm to 42 ± 8 Nm. Knee extensor peak torque improved slightly (but not significantly) in both legs. In the dominant leg, extensor peak torque improved from 54 ± 9 Nm to 56 ± 13 Nm, and in the nondominant leg, from 54 ± 10 Nm to 57 ± 15 Nm.

Statistically significant improvements were found in the hamstrings-to-quadriceps ratio for the nondominant leg, which increased from 71 ± 15% before training to 77 ± 16% after training (P = .01). There was no significant increase in this ratio in the dominant leg. No correlations were found between knee extensor and flexor strength and normalized knee and ankle separation distances for any phase of the jump-land sequence before or after training.

Implication of Findings

The authors developed and studied a videographic drop-jump screening test to record lower limb axial alignment in the coronal plane using a single standard videographic camera. The testing procedure was designed to be relatively easy to perform by researchers, coaches, trainers, or therapists in any facility. The goal was to provide a general indicator of an athlete’s lower limb axial alignment in the coronal plane in a straightforward drop-jump and vertical take-off task. The authors do not propose that this videographic analysis can be used as a risk indicator for a knee ligament injury. The test depicts only hip, knee, and ankle positions in a single plane during one maneuver, and it is recognized that noncontact ACL injuries may occur in side-to-side or cutting motions. A more sophisticated multicamera system would be required to measure these types of motions. The use of even just a second camera to measure the degrees of knee flexion on landing would provide additional information⁴⁴; however, the goal was to use one camera to make the measurements easier to perform from a screening standpoint. Because of these limitations, it may be that the single-plane videographic drop-jump test was not sensitive enough to depict relevant differences in landing patterns between the sexes.

Authors have reported that 58% to 61% of noncontact injuries occur on landing from a jump.^28,30 The final position of the knee joint on landing is influenced by the center of gravity of the upper body and the trunk over the lower extremity. Equally important are the positions of trunk-hip adduction or abduction, foot-ankle pronation-supination, and foot separation distance. These effects combine together to produce either a varus or a valgus moment about the knee joint that must be balanced by the lower limb musculature (see Fig. 16-1).⁴² If the athlete is off-balance, or has contact with another player, a loss of trunk and lower limb control and position may occur and the knee joint may go into a hyperextended or severe valgus position. The knee position measured in the coronal plane in the authors’ single-plane videographic analysis is a reflection of rotations in both the coronal and the transverse planes (internal-external femoral and tibial rotation). This test does not distinguish between these individual motions, which requires more sophisticated instrumentation.⁴²

Although there are many potential mechanisms for ACL injury, it has been postulated that excessive valgus moments about the knee joint with subsequent high anterior tibial shear forces may be one of the ACL injury mechanisms in females.^33,46,53 Excessive valgus loading may result in decreased tibiofemoral contact or condylar lift-off⁵⁴ and a reduction in the normal joint contact geometry that contributes to knee joint stability. This position, coupled with a highly activated quadriceps muscle that produces maximum anterior shear forces with the knee joint at low flexion angles (≤30°),^23,35,67 can lead to ACL rupture.

Hamstrings muscle activation with the knee joint at or near full knee extension (≤30°) produces insufficient posterior tibia shear forces to protect the ACL owing to the small angle of inclination of the hamstring tendons.⁷³ Noyes and Sonstegard⁶⁶ described a decreased mechanical advantage of the inner hamstrings when the knee is extended, because the hamstring flexion force at 0° is approximately half that at 90°. Similarly, the rotation force of the inner hamstrings declines with knee extension, because the rotation force at 0° is 41% that of the rotation force that occurs at 90°. A rotation “wind-up” effect exists, in which the IR forces increase significantly with ER owing to an increase in muscle mechanical advantage.

The authors hypothesized that the majority of untrained female athletes would demonstrate an abnormal lower limb position on landing from a drop-jump and on acceleration into a vertical jump. A normalized knee separation distance of less than 60% occurred in 63% of the untrained females on landing and in 80% on take-off. It was also hypothesized that the majority of young male athletes would demonstrate a neutrally aligned lower limb position on landing and acceleration into a vertical jump. This hypothesis was not supported, because 75% of the males had marked knee separation distances on landing and 72% on take-off, indicating a valgus-aligned lower limb position. The significance of this finding is unknown and may represent an aberration requiring further study (Fig. 16-10). It may be that this single-plane videographic test was not sensitive enough to depict relevant differences in landing patterns between the sexes. Other studies that used multicamera systems or force plates to investigate the drop-jump test found significant differences between females and males regarding knee flexion angles,⁴⁴ knee extension moments,^16,42 knee valgus moments,¹⁶ and ground reaction forces.⁴² A prior investigation determined that untrained female athletes had lower knee extension moments than males and postulated that this was explained by the male subjects’ high use of the hamstrings as a knee flexor at landing.⁴⁴

FIGURE 16-10 The drop-jump take-off sequences from three male athletes. A, Athlete 1: A 12-year-old soccer player demonstrates poor knee separation distance of 10.6 cm (28% normalized). B, Athlete 2: A 17-year-old basketball player demonstrated 29.3 cm (65% normalized) knee separation distance. C, Athlete 3: A neutral alignment is evident in this 16-year-old football player, who demonstrated 38.4 cm (82% normalized) knee separation distance.

(A–C, From Noyes, F. R.; Barber-Westin, S. D.; Fleckenstein, C.; et al.: The drop-jump screening test: difference in lower limb control by gender and effect of neuromuscular training in female athletes. Am J Sports Med 33:197–207, 2005.)

The authors believed that a neuromuscular-training program would increase knee separation distance in female athletes, thereby leading to a more neutral overall lower limb position during the drop-jump test (Fig. 16-11). The goals of the training program were to teach athletes to control the upper body, trunk, and lower body position; lower the center of gravity by increasing hip and knee flexion on landing; and develop muscular strength and techniques to land with decreased ground reaction forces. In addition, athletes were taught to pre-position the body and lower extremity before landing in order to land in the position of greatest knee joint stability and stiffness, with the goal of obtaining ankle and knee separation distances within approximately 80% that of the hip separation distance. The neuromuscular-training program was previously shown to be effective in inducing changes in neuromuscular indices (see Chapter 19, Decreasing the Risk of ACL Injuries in Female Athletes),⁴² including decreasing peak landing forces and peak adduction and abduction moments and improving hamstrings-to-quadriceps ratios. A correlation was found between the ability of the athlete to control knee adduction and abduction moments and the peak landing forces (P = .006). Decreased abduction or abduction moments may lessen the risk of lift-off of the medial or lateral femoral condyle, improve tibiofemoral contact stabilizing forces,⁵⁴ and decrease the risk of ligament injury. Other training techniques that use verbal and visual feedback have also been successful in reducing ground reaction forces.^{25,49,57,69,76}

FIGURE 16-11 The drop-jump take-off sequences from three female athletes before and after neuromuscular training. Athlete A, A 14-year-old basketball player demonstrated marked improvement in knee separation distance (from 17 cm to 37 cm) and in ankle separation distance (from 24 cm to 36 cm) after training (right). The normalized knee separation distance improved from 47% to 92%, and the normalized ankle separation distance improved from 64% to 92%. Athlete B, A 14-year-old volleyball player demonstrated poor upper body position and no improvement in knee or ankle separation distance after training (right). She was encouraged to pursue additional training.

In the female athletes, significant increases were found after training in the absolute and normalized knee and ankle separation distances for all phases of the jump-land sequence (P < .001). As a result, the female athletes had significantly greater normalized knee and ankle separation distances than the males for each phase. The neuromuscular-training program was reported to be effective in another study in decreasing the incidence of noncontact serious knee ligament injuries in female athletes.⁴⁰ In that report, 8 of 463 untrained females sustained a serious noncontact knee ligament injury compared with 0 of 366 trained females and 1 of 434 untrained male athletes. Untrained females had a 3.6 times higher incidence of knee injury than trained females (P = .05) and 4.8 times higher than males (P = .03).

It is unknown whether the improvement in knee separation distance after training will transfer to reduce the risk of a serious knee ligament injury in female athletes. The increased incidence of knee ligament injuries in female athletes is multifactorial and it is currently unknown which factors are dominant and which play a smaller role. A number of intrinsic factors inherent in women have been suggested, including a narrow intercondylar notch, smaller-sized ACL, pelvic-hip-knee-foot alignment, generalized knee laxity, foot pronation, and hormonal fluctuations.³³ Extrinsic factors related to athletic conditioning, skill, training, and equipment have also been discussed. Although issues related to differences in neuromuscular control, muscle reaction patterns,^45,93 and coordination and control of body and lower extremity positions during athletics are usually related to extrinsic factors, it may be that these factors are both intrinsic and extrinsic in their development.

It is important to note that some female athletes may require further training to demonstrate improvements in knee and ankle separation distance and land in a more neutrally aligned position. The pictorial sequence of the filmed drop-jump provides important information to athletes, parents, and coaches and can be useful in detecting athletes who require further neuromuscular training (see Fig. 16-11).

Hypothesis #1: No Significant Difference Exists between Prepubescent Boys and Prepubescent Girls in Overall Lower Limb Alignment during a Drop-Jump Test

The hypothesis was supported, because a marked decrease in normalized knee separation distance of less than 60% was found in 19 males (76%) and 25 females (93%) on landing (Fig. 16-12).

FIGURE 16-12 The distribution of male and female athletes according to normalized knee separation distance on landing. These percentile groups were chosen arbitrarily. There was no difference in the distribution of athletes among the categories shown. The stick figures are representative of the knee separation distance but not of the ankle separation distance.

(From Barber-Westin, S. D.; Galloway, M.; Noyes, F. R.; et al.: Assessment of lower limb neuromuscular control in prepubescent athletes. Am J Sports Med 33:1853–1860, 2005.)

There was no difference between the female and the male athletes in the mean centimeters of hip separation for each phase of the jump-land sequence. The mean hip separation distance was 36 ± 3 cm on pre-land and 35 ± 3 cm on landing and take-off for both genders. There was no association between normalized knee and normalized ankle separation distances for either gender on pre-land, landing, or take-off.

The male athletes had greater absolute ankle separation distances than the females (pre-land, 35 ± 5 cm and 28 ± 5 cm, respectively; landing, 32 ± 6 cm and 24 ± 6 cm, respectively; take-off, 32 ± 6 cm and 24 ± 5, respectively; P < .001), and greater normalized ankle separation distances as shown in Table 16-3. A wider 95% confidence interval (CI) range was noted for the male athletes than for the female values for all three phases of the test.

Male athletes also had greater absolute knee separation distances on pre-land and landing than females (pre-land, 22 ± 4 cm and 17 ± 4 cm, respectively; landing, 18 ± 7 cm and 14 ± 5 cm, respectively; P < .05), and greater normalized knee separation distances in these phases, as shown in Table 16-4. A wider 95% CI range was found for the male athletes on landing and take-off than for the females.

Hypothesis #2: No Significant Difference Exists between Prepubescent Boys and Prepubescent Girls in Knee Extensor and Flexor Strength

The hypothesis was supported, because there was no difference between genders in knee extensor and flexor peak torques (Table 16-5). No correlations were found between extensor and flexor peak torque values and overall lower limb alignment in the three phases of the drop-jump test. There was also no correlation between the isokinetic indices and BW or BMI for either gender.

Hypothesis #3: No Significant Difference Exists between Prepubescent Boys and Prepubescent Girls in Lower Limb Symmetry on Single-Leg Function Testing

The hypothesis was supported, because lower limb asymmetry (<85%) was found in 77% of the females and 84% of the males in the timed side-hop test, and in 55% of the females and 56% of the males in the triple-hop test. There was no difference between genders in the mean limb symmetry values for the timed side-hop test (68% ± 22% and 74% ± 22%, respectively) or the crossover hop for distance test (78% ± 21% and 76% ± 21%, respectively). There were no correlations between limb symmetry and knee isokinetic extension and flexion peak torques for either gender.

Implication of Findings

The authors hypothesized that no difference exists between genders in overall lower limb alignment during a drop-jump test. A valgus lower limb alignment assumed during athletic maneuvers has been hypothesized to increase the risk for a serious knee ligament injury, based on studies involving adolescent and adult athletes.^33,46 It was believed that prepubescent children would not demonstrate these gender differences because their muscle strength and motor skills had not yet fully developed.

The data indicated that 76% of boys and 93% of girls had a marked valgus alignment (<60% knee separation distance) during the videographic drop-jump test. This valgus lower limb alignment occurred even though the boys had greater ankle and knee separation distances throughout the drop-jump sequence. For example, the boys had a mean knee separation distance on landing of 49% ± 18% (95% CI, 42%–56%) compared with the girls who had a mean knee separation distance of 41% ± 12% (95% CI, 36%–46%). The mean normalized ankle separation distance of the males of 92% on landing has a theoretical advantage in providing a broader base (Fig. 16-13). However, the majority of boys still showed an overall lower extremity valgus alignment on landing and take-off, presumably owing to hip adduction and internal rotation. The predominance of this posture in the subjects would be expected to lead them to a higher injury risk. However, this risk may be offset in children by their smaller stature and body mass and the relatively slower velocities involved in their sporting activities.

FIGURE 16-13 The videographic drop-jump test photographs for the toe-touch and take-off phases are shown for two subjects. A and B, A 10-year-old female subject demonstrated poor knee separation distance in both pre-land (11 cm, 31% normalized) and take-off (12 cm, 33% normalized) phases.

Whether neuromuscular training would be effective in improving overall lower limb alignment to a more neutral position in children on a drop-jump test is unknown. Few controlled studies have been conducted to investigate injury-prevention strategies in children.^14,52,59 There are no formal national education or certification requirements for coaches in the United States, and it is unknown how many coaches have had formal training in conditioning and training, growth and development, or injury prevention.⁵⁹ The continued analysis of knee ligament injury mechanisms is strongly advised, because newer coaching techniques and strategies may become effective in addressing the ACL injury problem.

The finding of a lack of a difference in all of the isokinetic muscle strength values between the boys and the girls agrees with other studies in children of similar ages.^21,22,86 Gender-related strength differences are not expected to become apparent until the mid-teens.^{21,22,38,43,79,83} One purpose of conducting isokinetic testing in the study was to assess whether a correlation existed between knee extensor and flexor strength and knee or ankle separation distances, of which none was found.

The authors hypothesized that there would be no difference between boys and girls in lower limb symmetry as determined by single-leg-hop tests. One previous study assessed the reliability of two sports-related functional tests that involved speed and agility during double-legged activities (slalom and hurdle tests) in 11 athletes (8 boys and 3 girls) who were all 11 years of age.² The intraclass correlation coefficients (ICCs) demonstrated adequate reliability of these tests. The standing balance of children 10 years of age and younger has been shown to be poorer than that of adults owing to their developing vestibular system.¹⁷ Postural sway has been shown to be greater in children 9 to 10 years of age than in those 15 to 18 years of age.⁹⁵ In our study, abnormal limb symmetry was found in 77% of the females and 84% of the males in the timed side-hop test and in 55% of the females and 56% of the males in the triple-hop test. The timed side-hop test induced medial and lateral forces, and the children had a difficult time maintaining balance on one leg while hopping in these directions.

There was no correlation between knee extensor and flexor strength and lower limb symmetry on the single-leg-hop tests. Prior studies have shown a relationship between extensor strength and single-leg-hop testing in older populations 16 to 48 years of age.^6,63 The results of the functional tests could be due to motor development skills and neuromuscular indices such as balance, proprioception, and hip muscle strength not analyzed in this study. Cross-sectional investigations using single-leg-hop tests on children and adolescents up to 18 years of age are necessary to determine whether an age effect exists similar to those that exist for isokinetic muscle strength¹⁹ and standing balance.¹⁷

The overall results of this investigation showed that a high percentage of male and female athletes 9 to 10 years of age had a noteworthy valgus lower limb alignment during a drop-jump screening test and a lack of symmetry between lower limbs in single-leg-hop tests. These same indices have been hypothesized to increase the risk for serious knee ligament injuries in adolescent and adult athletes. Consideration should be given to incorporate neuromuscular-training techniques that educate proper landing mechanisms and body positioning, and train to improve balance and muscle strength, into younger athletes.

Hypothesis #1: A Significant Increase in Isokinetic Knee Extensor and Flexor Strength Occurs with Age in Both Male and Female Athletes

This hypothesis was supported for knee extensor strength in both female and male athletes. In the female athletes, a significant increase in the normalized mean extensor peak torque was noted from the ages of 9 to 13 (P < .001; Fig. 16-14). The average extensor peak torque in the 13-year-olds was 20% greater than that of the 9-year-old subjects. There were no significant increases in this factor after the age of 13. In the male athletes, significant increases in mean extensor peak torque were found from ages 9 to 14 (P < .001). The average extensor peak torque in the 14-year-olds was 38% greater than that of the 9-year-old subjects. There were no significant differences in this factor after the age of 14.

FIGURE 16-14 No significant difference was found in the mean extensor peak torque ratio values between females and males in the 9- to 13-year-old age groups. However, from the ages of 14 to 17, males had significantly greater mean extensor peak torque ratio values than females.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

Whereas this hypothesis was supported for isokinetic knee flexion strength in male athletes, it was not supported in the female athletes. The normalized mean flexor peak torque was only slightly greater in female athletes 11 years of age than in those 9 to 10 years of age (Fig. 16-15). There was no significant difference in flexor peak torques in athletes older than the age of 11. Hamstrings-to-quadriceps ratios slightly declined with age (from 82% at age 10 to 70% at age 17). In the male athletes, normalized mean flexor peak torque significantly increased from 80 ± 26 Nm/BW at age 9 to 104 ± 16 Nm/BW at age 14 (P < .001). Hamstrings-to-quadriceps ratios declined significantly from ages 9 (mean, 87%) to 14 (mean, 72%; P < .001), after which this value remained constant. There was no correlation between age and extensor or flexor peak torque absolute values, peak torque normalized values, or the hamstrings-to-quadriceps ratio for either gender.

FIGURE 16-15 No significant difference was found in the mean flexor peak torque ratio values between females and males in the 9- to 13-year-old age groups. However, from the ages of 14 to 17, males had significantly greater mean flexor peak torque ratio values than females.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

Hypothesis #2: Males Aged 14 and Older Have Significantly Greater Lower Limb Muscle Strength than Age-matched Females

This hypothesis was supported, because from ages 14 to 17, male athletes had significantly greater mean extensor and flexor peak torques than age-matched female athletes (Table 16-7). There was no significant difference between genders in the hamstrings-to-quadriceps ratio in any age group.

Hypothesis #3: Age and Gender Do not Influence Lower Limb Symmetry on Single-Leg-Hop Functional Testing

The hypothesis was supported, because no statistically significant difference was found in the mean lower limb symmetry values between any of the age groups for either the female or the male athletes in the single-leg-hop tests (Figs. 16-16 and 16-17). There was no correlation between extensor and flexor peak torques and limb symmetry.

FIGURE 16-16 There was no effect of age or gender on limb symmetry in the timed side-hop test.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

FIGURE 16-17 There was no effect of age on limb symmetry for either gender in the crossover hop for distance test. Males 15 years of age had greater mean limb symmetry values than age-matched females.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

There were no significant differences between male and female athletes in the single-leg-hop tests, except in the crossover hop for distance test in which the 15-year-old males had a greater mean limb symmetry value than the females (P < .05).

Hypothesis #4: Age and Gender Do not Influence Overall Lower Limb Alignment on a Drop-Jump Test

The hypothesis was supported, because there was no effect of age found in either the female or the male athletes on the mean normalized ankle and knee separation distances on landing (Fig. 16-18) or take-off. There was no significant difference in the percentage of athletes within the categories of less than 60%, 61% to 80%, or greater than 80% normalized knee separation distance (Fig. 16-19). A distinct overall lower limb valgus alignment (<60% knee separation distance) was measured in 78% of female athletes aged 9 to 12, 81% of athletes aged 13, 83% of athletes aged 14, 71% of athletes aged 15, and 74% of athletes aged 16 to 17. In the male athletes, a distinct valgus alignment was found in 79% of those aged 9 to 12, 70% of athletes aged 13, 62% of athletes aged 14, 67% of athletes aged 15, and 80% of athletes aged 16 to 17.

FIGURE 16-18 Normalized ankle and knee separation distances on landing during a videographic drop-jump test in 396 female and 140 male athletes. There was no effect of age on these data in the female athletes. An age effect was found in the male athletes in the mean ankle separation distance, but not in the mean knee separation distance.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

FIGURE 16-19 There was no significant difference between either the male or the female athletes aged 14 to 17 in the percentage of subjects with <60%, 61%–80%, or >80% normalized knee separation distance. The stick figures are representative of the knee separation distance but not of the ankle separation distance.

(From Barber-Westin, S. D.; Noyes, F. R.; Galloway, M.: Jump-land characteristics and muscle strength development in young athletes: a gender comparison of 1140 athletes 9 to 17 years of age. Am J Sports Med 34:375–384, 2006.)

Significant differences were found in the male athletes between the age groups in mean ankle separation distances for all three phases of the drop-jump test (P < .005). The ankle separation distances were greatest in the 9- to 12-year-olds and smallest in the 15-year-olds. Significant differences were found between genders in the mean ankle separation distances in the pre-land phase (Table 16-8). Although male athletes had greater mean ankle separation distances in the 9- to 12-year-old group, female athletes had greater mean values in the 13-, 15-, and 16- to 17-year old groups. Few significant differences were found between genders within the age categories in the mean normalized knee separation distances.

Study Implications

The first hypothesis was that a significant increase in lower limb muscle strength occurs with age in both male and female athletes. There is disparity in the literature regarding age effects on knee extensor and flexor strength within gender owing to the few comprehensive cross-sectional or longitudinal studies that have been published. No age effects were found in females or males when tested at a low-velocity speed (30°/sec) after stature and mass were accounted for in two investigations.^21,22 Other authors reported that an age effect existed in both genders for extensor and flexor torques at velocity speeds ranging from 12°/sec to 150°/sec.⁸⁶ Another report claimed that an age effect existed for males, but not for females, in knee extensor and flexor peak torques at 300°/sec.⁴¹ De Ste Croix and colleagues²² tested 46 subjects aged 8 to 9 years, 47 subjects aged 13 to 14 years, and 48 subjects aged 18 to 27 years. Knee extensor and flexor strength increased with age, regardless of how the data were analyzed (absolute, body mass–related, or body mass–adjusted). Ramos and coworkers⁷⁹ tested 18 subjects aged 11 to 12 years, 21 subjects aged 13 to 14 years, and 18 subjects aged 17 to 18 years, and reported increases in absolute and BW-corrected peak torque with age in both males and females.

In the authors’ study, an increase in knee extensor peak torque of 20% was observed in the female subjects between the ages of 9 and 13. However, only a small (nonsignificant) mean increase of 16% in knee flexor peak torque was measured between the ages of 9 to 11, with no significant improvement noted in athletes older than 11 years. In the male subjects, an increase in mean knee extensor peak torque of 38% was found between the ages of 9 and 14 and an increase in mean knee flexor peak torque of 23% from the ages of 9 and 14.

The second hypothesis was that males aged 14 and older have significantly greater lower limb muscle strength than age-matched females. Three investigations on isokinetic testing in children reported no differences between boys and girls in knee extensor or flexor strength up to the age of 14, even after controlling for body mass and stature.^21,22,79 Our data was in agreement, because there was no difference between genders in lower extremity muscle strength until the age of 14. One of the most important findings from this investigation was the difference between the male and the female athletes in strength increases in knee flexor peak torque according to the age categories. Whereas flexor peak torques in males gradually increased from ages 9 to 14, such increases were noted in females only from ages 9 to 11. In addition, the increases noted in extensor and flexor peak torques with age in males was approximately double those of females, even after controlling for BW. One theory of the increased incidence of ACL injury in female athletes compared with males relates to the quadriceps-dominance effect, or greater quadriceps activation measured during functional activities.^16,53 An increased quadriceps muscle force combined with a decreased hamstring muscle force and decreased knee flexion angle produce an increased proximal tibial anterior shear force that may result in excessive strain on the ACL.^16,45,53,80 It would appear from this study that an emphasis on hamstring-strengthening programs needs to be further pursued in young female athletes to enable them to continue to achieve strength gains past the age of 11.

The third hypothesis was that age and gender do not influence lower limb symmetry on single-leg-hop functional testing. There was no difference in limb symmetry between the age categories in either the male or the female athletes. Although the mean values of the two hop tests indicated adequate limb symmetry was present within most age categories, the large standard deviations prevent definitive conclusions. In a previous study from the authors’ institution that involved 93 subjects aged 17 to 34 years, 90% had limb symmetry of at least 85%. A relationship between extensor strength and single-leg-hop testing was reported in one study⁶; however, a second investigation failed to find a significant correlation.⁶³ No association was found in the current investigation between extensor or flexor strength and limb symmetry values. Studies conducted elsewhere have presented wide variability in regard to a correlation between extensor strength and limb symmetry on single-leg-hop testing; moderate correlation coefficients (R = 0.60–0.78) were reported by Greenberger and Paterno,³¹ Sachs and associates,⁸¹ Wilk and colleagues,⁹⁰ and Barber and coworkers.⁶ However, low correlation coefficients were described by other investigators,^71,74,75,84 and no definitive conclusion may be reached regarding this association.

The fourth hypothesis was that age and gender do not influence overall lower limb alignment on a drop-jump test. Some authors have postulated that female athletes tend to land from a jump in a noteworthy valgus lower limb alignment, in contrast to male athletes who land in a more neutrally aligned position, and that a valgus alignment position increases the risk for a knee ligament injury.^33,46 However, based on data from prior investigations, we believed no difference in lower limb alignment on a drop-jump task would be detected between genders.^7,65 The results showed that, although males aged 9 to 12 had greater mean ankle separation distances than age-matched females, and males aged 14 had greater knee separation distances on take-off than females, the majority of all athletes had a distinct valgus lower limb alignment on landing. Valgus knee alignment on take-off and landing has been hypothesized to predispose an athlete to ACL rupture and indeed is commonly seen during noncontact ACL injuries.^11,28,30

Hass and associates³⁷ recently reported an age-dependent effect on knee biomechanics during landing between 16 prepubescent (aged 8–11 yr) and 16 postpubescent (aged 18–25 yr) females. The postpubescent group had reduced knee flexion upon initial contact (4.5°) and increased medial knee forces, but decreased landing forces and knee extensor moments than the younger group. Although the study’s results suggested the presence of a developmental effect on landing mechanisms, the authors concluded that further work is required on the effects of age, developmental level, and motor control during landing.

Chappell and colleagues¹⁶ reported no difference between genders aged 19 to 25 years in the magnitude of knee varus-valgus moments during three stop-jump tasks. Those authors concluded that the female knee varus-valgus moment may not be responsible for the gender differences in ACL strain during those tasks. The increased incidence of knee injuries in female athletes is most likely due to many factors. Whereas a valgus alignment position assumed on landing may increase the risk for a knee ligament injury, the fact that there was no difference detected in alignment between genders in the authors’ study in such a large number of athletes indicates that other factors most likely have a role in the increased ACL injury rate in females. In addition, injury mechanisms other than a valgus knee position have been noted during ACL ruptures,¹¹ including a sudden deceleration, acceleration, or change of direction in response to avoiding contact with another player or in reaction to a ball or play. Differences between genders in muscle reaction and firing patterns,^45,93 coordination, control of body and lower extremity positions during athletics, lower extremity and hip strength, and external-internal knee flexion moments all most likely contribute to this problem. Neuromuscular training (see Chapter 19, Decreasing the Risk of ACL Injuries in Female Athletes) is effective in improving lower limb alignment on the drop-jump test,⁶⁵ increasing hamstrings strength, increasing knee flexion angles on landing,^42,44,76 and reducing deleterious moments and ground reaction forces.

Hypothesis #1: A Significant Increase in IR and ER Strength Occurs with Age in Both Male and Female Athletes

The hypothesis was proved true for the male athletes, but not for the female athletes. Male athletes aged 14 to 17 years had significantly greater mean IR and ER peak torques (Table 16-9), and significantly greater extensor and flexor peak torques (Table 16-10), than males aged 11 to 13 years. There was no significant difference in all of the lower limb muscle strength indices in female athletes between the ages of 11 to 13 years and 14 to 17 years.

There were no age effects for either gender on the IE ratio. Only one age effect was found for the rotation-flexion and rotation-extension ratios: females aged 11 to 13 years had a significantly greater mean IR-E ratio than females aged 14 to 17 years.

Hypothesis #2: No Difference Exists between Male and Female Athletes Aged 11 to 13 in IR and ER Strength

The hypothesis was supported, because there was no significant difference between male and female athletes 11 to 13 years of age regarding IR and ER strength (Fig. 16-20). Females aged 11 to 13 years had significantly greater mean IR-F and IR-E ratios than age-matched males.

FIGURE 16-20 Overall isokinetic peak torque averages for knee extensor, flexor, external rotation, and internal rotation for the dominant limb at 180°/sec. A hierarchy of isokinetic muscle strength about the knee can be seen, which is similar between genders and age groups.

(Reprinted from Noyes, F. R.; Barber-Westin, S. D.: Isokinetic profile and differences in tibial rotation strength between male and female athletes 11 to 17 years of age. Isok Exer Sci 13:251–259, 2005; with permission from IOS Press.)

Hypothesis #3: Male Athletes Aged 14 to 17 Have Significantly Greater IR and ER Strength than Age-matched Female Athletes

The hypothesis was supported for IR only, because males aged 14 to 17 years had a significantly greater mean IR peak torque (at 180°/sec) and a significantly faster mean time-to-reach IR peak torque than age-matched females (see Table 16-9). There was no significant difference between the male and the female athletes for mean ER peak torque.

In a subgroup of 14- to 17-year-old athletes matched for BMI, males had significantly greater IR peak torques at 120°/sec than females (P < .05). Males in this subgroup also had significantly greater knee extensor (P < .01) and flexor peak torques (P = .0001) than females.

Study Implications

This is the first study to conduct lower extremity tibial rotational strength testing in athletes 11 to 17 years of age with the specific purposes of assessing age- and gender-related effects. During a potential injury situation, dynamic muscular stabilization of the knee joint, particularly the hamstring muscles, may potentially reduce ACL strain.^{24,45,61,77,80} Several investigators have demonstrated that female athletes have weaker knee flexor peak torques, and take longer to develop flexor peak torque, than males even after normalizing for BW.^18,32,42,45 The impetus for this study was the paucity of published data regarding the IR and ER isokinetic strength of the hamstring musculature.

The isokinetic test design and protocol was performed in a manner similar to that of other investigations.^39,82,89 The joint angular positions were based on biomechanical studies that demonstrated that the rotational force of the pes anserinus musculature (sartorius, gracilis, semitendinosus) is greatest at 90° of knee flexion.^66,85 A significant decrease in rotational force exists as the knee is extended, declining as much as 41% when the knee is fully extended (0°) compared with the strength measured at 90° of flexion. Hip flexion angle also affects the generation of internal rotational torque, with greater torque generated at higher hip flexion angles (when the knee is flexed 90°).^70,85 The test protocol was similar to that described by other investigators with respect to hip and knee flexion angles, stabilization of the foot, and the resting position used to determine the neutral position before the onset of testing.^39,82,89

A limitation of this study is the fact that it is not possible to isolate and measure the specific muscles that act as the primary internal rotators (semitendinosus, gracilis, popliteus, semimembranosus, and sartorius) and external rotators (biceps muscle, long and short head)¹³ with dynamometer systems. However, the testing procedure does provide a functional assessment of lower limb isokinetic capabilities. The data showed few effects of lower limb dominance on rotational strength, which agreed with the results of Osternig and coworkers⁷² and Hester and Falkel.³⁹ The results of the IE ratios were also similar with those reported previously, because the mean ER peak torque slightly exceeded that of IR peak torque at both velocities.^39,72

The first hypothesis, that a significant increase in IR and ER strength occurs with age in both genders, was proved for males but not for females. Male athletes aged 14 to 17 years had significantly greater mean IR (17% for both test velocities) and ER (14%–18%) peak torques than males aged 11 to 13 years. In contrast, females aged 14 to 17 years had only 6% greater ER strength than that measured in 11- to 13-year-old females. Further rotational strength testing is required in cross-sectional or longitudinal studies before a conclusion can be reached on the effects of aging on IR and ER peak torques. Our study did demonstrate a hierarchy of isokinetic peak torques in the musculature about the knee, which was similar in both genders and age groups (see Fig. 16-20).

Our second hypothesis, that no difference exists between genders aged 11 to 13 years in IR and ER strength, was supported. Other investigators have successfully used a variety of isokinetic velocity speeds to measure knee extension and flexion peak torque in children, ranging from 12°/sec⁸⁶ to 300°/sec.²⁶ No significant gender-related knee flexion and extension strength differences have been reported by authors in children between the ages of 8 and 14 years.^21,22,86 Our investigation showed no difference between genders aged 11 to 13 years in mean ER and IR peak torque values, time-to-peak torque, or IE ratios.

Our third hypothesis, that male athletes aged 14 to 17 years have significantly greater tibial rotation isokinetic muscle strength than age-matched female athletes, was supported. The IR peak torques for the males were 17% greater than those of the females at 180°/sec. Viola and associates⁸⁹ reported that the noninjured limb IR of males was approximately 25% stronger than the uninjured limb IR of females at 120°/sec; and approximately 11% stronger at 180°/sec. These data were not normalized for BW. Segawa and associates⁸² reported that (after normalization for BW) the male noninjured limb IR was approximately 16% stronger than the female uninjured limb IR at 120°/sec. There appears to be no difference in ER strength between genders, according to the data from our study and those of other investigators.^89,93

Our study is the first we are aware of to report that male athletes have faster times-to-reach peak torque for IR than age-matched female athletes. Gender differences in muscular protection of the knee joint during weight-bearing and non–weight-bearing activities have been reported by many authors, but only a few have included IR data. Wojtys and colleagues⁹³ compared the amount of IR and rotational stiffness in collegiate athletes participating in high-risk sports with those participating in low-risk endurance sports. Gender comparisons were performed regarding the amount of passive and active IR in response to a medially directed 80-N force (with the subjects seated) and the resultant amount of apparent knee joint stiffness. Male athletes participating in high-risk sports had significantly less IR (mean, 3°) during muscle activation, and significantly greater (42%) apparent joint stiffness than the female athletes. However, Lephart and coworkers⁵⁰ found that collegiate women who participated in high-risk sports had less knee flexion and IR maximum angular displacements during single-leg landing and forward hop landing tasks than age- and activity-matched males. McLean and associates⁵⁵ reported that women had an average of 5° less IR during a sidestep maneuver, along with less hip flexion, hip abduction, hip internal rotation, and knee flexion than age-matched males. The protection provided by the dynamic neuromuscular control of the lower extremity is multifactorial and beyond the scope of the authors’ study.