Comparison of Drop Jumps and Sport-Specific Sidestep Cutting

Methods

All players of the Norwegian elite female handball series were invited to baseline testing for a cohort study to investigate ACL injury risk factors. A high ACL injury incidence has previously been found in this cohort. ¹³ We tested 184 players, and from 173 match-fit players the 125 back and wing players were selected for analysis, as they are most accustomed to sidestep cutting during match play. Four players were excluded because of technical problems and 1 excluded because of physical complaint during jumping. The final sample consisted of 120 players (mean ± SD: age, 22.4 ± 7.1 years; height, 171 ± 7 cm; weight, 67 ± 7 kg).

The study was approved by the Regional Ethics Committee and signed informed consent forms were obtained from all players.

Sidestep cutting and drop jumps were performed in a motion analysis laboratory with eight 240-Hz infrared cameras (ProReflex, Qualisys, Gothenburg, Sweden) and two 960-Hz force platforms (AMTI, Watertown, Massachusetts). Markers were placed and sidestep cutting procedure was performed as described in a previous study from this cohort. ¹⁰ The players were told to perform their regular sidestep cut, trying to fake a static defender into going one way while cutting to the other side. They received a pass before cutting (Figure 1).

OPEN IN VIEWER

Figure 1.

Testing situation. The approach of the players was approximately 33° on the long axis of the runway. Their instruction was to fake the defender into going to one side and cut to the other. The defender was static during recording and adjusted her position between the trials to make sure the athletes hit the force platform with their normal sidestepping technique. (Reprinted from Kristianslund E, Krosshaug T, van den Bogert AJ. Effect of low pass filtering on joint moments from inverse dynamics: implications for injury prevention. J Biomech. 2012;45:666-671. With permission from Elsevier.)

Drop jumps were conducted using a 30-cm box. The subjects were instructed to drop off the box and perform a maximal jump after landing. The box was adjusted so the players landed with 1 foot on each platform. Static recordings of the athlete in an anatomically neutral position were performed before testing. Sidestep cutting to both sides was completed before the jumping trials.

The contact phase was defined as the period where the unfiltered vertical ground-reaction force exceeded 20N. Kinematics and kinetics were calculated as previously described, ¹⁰ except from the hip joint center that was calculated by the regression equations of Bell et al. ¹ A 15-Hz cutoff frequency was applied for signal filtering of both force and position data. External joint moments projected onto the knee rotational axes are reported. Both knees were analyzed in drop jumps, whereas the right knee was analyzed in right-left sidestep cuts and the left knee in left-right cuts.

Three drop jumps and 3 sidestep cuts from each side were selected for analysis.

Statistical Treatment

The following variables were extracted from the motion analysis of drop jumps and sidestep cuts: maximum knee abduction and knee internal rotation moments in the first 100 milliseconds after initial contact; maximum knee flexion moment during contact with force platform; knee flexion, knee abduction, and knee internal rotation at initial contact (IC); and maximum knee flexion, knee abduction, and knee internal rotation. Maximum knee abduction and knee internal rotation moments during the first 100 milliseconds were chosen because ACL injuries are likely to occur in this period. ⁹ Average values over 3 trials for each knee were used for analysis, and sidestep cutting and drop jumps were compared for each knee. The Spearman rank correlation coefficient (ρ) was calculated between knee abduction moments and knee valgus angles in jumps and sidestep cuts to assess if there was a difference of ranking of players based on these parameters between the tasks. The correlation between knee valgus angles and knee abduction moments in drop jumps and sidestep cuts as well as the correlation between knee valgus angles in drop jumps and knee abduction moments in sidestep cutting were also found using the Spearman rank correlation.

Results

We observed a peak in knee abduction moments shortly after initial contact in sidestep cuts but not in drop jumps (Figure 2). The sidestep cuts were performed with mean approach speed of 3.4 m/s, and the cutting angle was 69°. Knee joint angles (Table 1) and knee joint moments (Figure 3) were substantially different between jumps and sidestep cuts. The Spearman ρ was 0.135 for knee abduction moments (Figure 4) and 0.706 for knee valgus angles (Figure 5), indicating poor and moderate agreement between tasks, respectively. There was limited correlation between valgus angles and knee abduction moments within each of the movement tasks. In the drop jumps, we observed a rank correlation of 0.506, whereas the rank correlation for sidestep cuts was 0.339. The rank correlation between knee valgus angles in drop jumps and knee abduction moments in sidestep cuts was poor (ρ = 0.238).

OPEN IN VIEWER

Figure 2.

Knee abduction moments (mean ± SD) during the first 150 ms of stance in jumps and sidestep cuts. Both knees, N = 720 trials.

OPEN IN VIEWER

Table 1

Knee Joint Angles and Moments (in Degrees) in Vertical Drop Jumps and Sidestep Cuts (N = 240 Knees)

	Jumps			Sidestep Cuts
	Mean	SD	95% CI	Mean	SD	95% CI
Flexion at initial contact (IC)	31.5	6.48	30.70, 32.35	20.9	5.37	24.52, 25.95
Valgus at IC	−1.2	4.03	−1.75, −0.72	4.6	3.81	4.11, 5.08
Internal rotation at IC	−1.4	5.95	−2.12, −0.60	2.0	7.59	1.01, 2.94
Maximum flexion	82.2	11.76	80.70, 83.69	62.2	5.10	61.57, 62.86
Maximum valgus	5.6	4.63	4.98, 6.16	11.5	4.94	10.82, 12.08
Maximum internal rotation	9.3	5.25	8.63, 9.96	12.6	5.22	11.98, 13.31

OPEN IN VIEWER

Figure 3.

Maximal knee joint moments during the first 100 ms of stance. N = 240 knees.

OPEN IN VIEWER

Figure 4.

Scatter plot of maximum knee abduction moment (in Nm/kg) during the first 100 ms in jumps and sidestep cuts. N = 240 knees. Lines at mean + 1 SD.

OPEN IN VIEWER

Figure 5.

Scatter plot of maximum knee valgus angle (in degrees) during stance phase in jumps and sidestep cuts. N = 240 knees. Lines at mean + 1 SD.

Discussion

There was a substantial difference in kinematics and kinetics between drop jumps and sport-specific sidestep cutting, as previously reported by others.^3,4,14,15 In sidestep cutting the athletes had lower knee flexion angles and higher knee valgus and internal rotation angles at IC and at maximum. The knee joint moments were higher in all 3 planes for the sidestepping movement. Most notably, the knee abduction moments were 6 times higher in sidestep cutting compared with drop jumps. Sidestep cutting is a high-energy situation with a high approach speed, direction change, and single-legged stance, compared with the more controlled double-leg drop jump, and this may explain the differences in kinematics and kinetics.

There was a weak correlation between knee abduction moments in drop jumps and sidestep cutting, while the correlation was better for the knee valgus angles. Abduction motion and loading are important components of the injury mechanism.^8,9,21 Unfortunately the correlation of drop jump knee valgus angles to sidestep cutting knee abduction moments was poor, hence visual drop jump tests cannot predict loading of the knee during the sidestep cutting maneuver in which a high proportion of ACL injuries occur.^11,16 However, in the prospective study of Hewett et al, ⁶ both knee valgus angles and knee abduction moments in drop jumps predicted ACL injuries among adolescents. The moderate correlation between knee valgus angle in drop jumps and sidestep cutting indicates that drop jump tests may have potential to identify athletes with high knee valgus angles during cutting activities and provides an opportunity for screening of motion patterns that are likely to be relevant to ACL injury causation.

Knee abduction moments may be less relevant for ACL injury risk in our cohort of elite female handball players, as the high knee abduction moments seen among injured players in the prospective study of Hewett et al ⁶ are not seen in our cohort. Our athletes are older, and players with such high knee abduction moments may have been injured or have other characteristics that have excluded them from elite level sports. In addition, the prospective study may have been affected by artifacts due to inconsistent filtering of force and movement data.^7,10,18

Injuries occur mostly during single-legged activities, and efforts should be made to find tests that can identify players with high knee valgus angles during sporting activities. The limited ability of a clinical drop jump test to predict ACL injury may indicate that more sport-specific tests are needed. Based on the present knowledge, all female team sports athletes should perform preventive exercises regularly, regardless of presumed injury risk.

A limitation of all laboratory studies is that one cannot conclude how the measured movement patterns relate to the biomechanics of real sporting motions. However, we attempted to simulate real sport-specific situations for sidestep cutting by including a static opponent and a ball, with observers continually assessing the intensity and sport-specific quality of the cuts. The players were specifically told to perform the cut as if they were trying to fake the static defender into going the opposite way. This requires a high intensity of the cut to trick the opponent. The loads calculated during this more sport-specific sidestep faking maneuver are likely closer to the loads experienced during game play than loads from analyses of simple changes of direction. The conclusions can likely be extrapolated to other sports, as faking a defender and cutting past him or her is very common in different team sports (eg, basketball and soccer).

A limitation of this cutting protocol is that it is harder to standardize, as all athletes use their preferred cutting technique. On the other hand, the resulting variation in sidestep cutting technique likely reflects the variation in cutting techniques used during active game play, and the high number of subjects ensures representative data. Sidestep faking and cutting maneuvers during active game play usually include an element of spontaneity. ² However, with inclusion of spontaneity the task would be less standardized.

Although this is a cross-sectional study with no data on actual injury risk, the findings are useful for screening for ACL injuries and development of ACL injury prevention programs. Knowledge of the relation between joint kinematics and kinetics in the drop jump screening test and in potential injury situations can help design better screening tests.

Conclusion

There was a substantial difference in the magnitude of knee joint deflections and knee joint loading between drop jumps and sport-specific sidestep cutting. There was a poor correlation between knee abduction moments between the tasks, indicating that the players with high frontal plane loading in drop jumps might not necessarily experience high frontal plane loads in sidestep cutting. The kinematics is more consistent across tasks than the kinetics and may be a more relevant target for ACL injury risk screening.