Journal Logo


Effects of Isolated Hip Abductor Fatigue on Frontal Plane Knee Mechanics


Author Information
Medicine & Science in Sports & Exercise: March 2010 - Volume 42 - Issue 3 - p 535-545
doi: 10.1249/MSS.0b013e3181b7b227
  • Free


Anterior cruciate ligament (ACL) injuries and patellofemoral pain syndrome (PFPS) both occur at significant rates in athletics, more often in college and high school age groups, and more frequently in women than in men (21,32). These injuries cost society financially both in immediate costs associated with acute medical treatment (7,21,25) and in future costs for treatment of associated comorbidities such as meniscal tears and knee osteoarthritis (11,28). Thus, substantial efforts have been made to understand the cause of these injuries and offer methods for their prevention.

Although the cause of each injury is considered multifactorial (22,40), movement patterns in the frontal plane have emerged as a risk factor for both injuries. Acute rapid increases in frontal plane knee abduction range of motion (ROM) have been demonstrated to increase strain on the ACL, at times enough to induce ligament rupture (8,13,19). This is in agreement with observational studies that report that most ACL injuries involve a mechanism that includes a valgus positioning at the knee (2,35). In addition, a prospective study demonstrated that an increase in knee abduction angle at initial contact with the ground and in the peak knee (external) abduction moment at the knee during the ground contact phase were predictive of increased risk for ACL injuries in a group of female soccer, basketball, and volleyball players (17). Thus, frontal plane knee movement may be an important factor in ACL injuries.

During repetitive movements, frontal plane loading may not reach the levels necessary to put the ACL at risk. However, increased knee abduction has been hypothesized to laterally displace the patella (36), which increases the compressive forces between the lateral facet and the lateral femoral condyle (8,26). This is hypothesized to lead to PFPS (36). Therefore, frontal plane motion and loading seem relevant to both the occurrence of ACL injuries and the development of PFPS.

It is possible that weakness of the hip abductors is responsible for the increases in knee abduction ROM and knee adductor moments because the hip abductors primarily stabilize the femur during frontal plane lower extremity motion (31). Bobbert and van Zandwijk (1) reported that in the sagittal plane, the moments at the knee and ankle were largely a function of the forces generated by the hip musculature. A computer model has demonstrated this in the frontal plane. A 50% decrease in hip abductor stiffness led to alterations in knee abduction motion and decreased the axial load that the knee could withstand before surpassing an injury threshold set at an estimated level of knee stiffness in the frontal plane (6). This was most apparent when the knee was positioned in greater valgus. In addition, hip abductor weakness has been associated cross-sectionally with several lower extremity injuries including iliotibial band friction syndrome (10), anterior knee pain/PFPS (23), and general overuse injuries (33). Brindle et al. (4) also identified decreased hip abductor activation in participants with PFPS while negotiating stairs. However, these studies leave in doubt whether hip abductor weakness, insufficiency, and decreased activation existed before these injuries or became evident secondary to the injury. The acute effects of decreased hip abductor force output on knee kinematics and kinetics have not been reported. Yet, many injury prevention programs now focus on conditioning and learning to control dynamic frontal plane knee movement (18), although the initial causes of altered frontal plane knee movement characteristics are not yet fully understood.

Previous studies have used a fatigue protocol to simulate insufficiency of a muscle group to evaluate the effects on movement characteristics (5,14,34). These studies provide a design framework with which to examine the effects of decreased muscular force production on movement characteristics. However, no previous study has precisely focused on the kinematic and kinetic changes in the lower extremity with isolated hip abductor fatigue-induced weakness during unilateral activity. Therefore, the purpose of this study was to identify the effects of isolated hip abductor fatigue-induced weakness on lower extremity kinematics and kinetics in recreationally active women. Specifically, we hypothesized that, after an isolated hip abductor fatigue protocol, internal hip abductor moments would decrease, and increases would be noted in hip adduction ROM, knee abduction ROM, and internal knee adductor moments during the critical weight acceptance (WA) phase of landing activities.


Institutional review board (IRB) approval was obtained for this study from the University of Wisconsin-Milwaukee IRB, and a deferral was obtained from the Marquette University IRB to the IRB at UW-Milwaukee, where the research was performed.


Data from previous published literature (5,17) were used to estimate the effect size of hip abductor fatigue or weakness on knee moment. From these data and from pilot testing, an effect size of d = 0.75 was estimated for the effect of hip abductor fatigue on knee frontal plane angle and moment. This effect size estimate, a desired power of 80% (1 − β = 0.80), and desired α = 0.05 were used to calculate the necessary sample size to be 17 participants. To account for participant attrition, 20 recreationally active female participants were recruited from the local university population. Participants were aged 20.7 ± 1.7 yr, 1.67 ± 0.06 m in height, and 63 ± 10.3 kg in weight. Participants were free from any lower extremity pain or injury that limited activity in the past 6 months and from any past injury requiring surgery to the lower extremity. They were involved in recreational or competitive physical activity involving ground impact, such as running, basketball, or aerobics, for at least 30 min three times per week. Informed consent was obtained from all subjects on arrival for the study, and all questions were answered.


Three-dimensional motion of the pelvis and dominant stance lower extremity was collected via a seven-camera Eagle Digital Camera System (Motion Analysis Corp, Santa Rosa, CA). Data were collected using the EVa Real Time software, version 5.0 (Motion Analysis Corp.). Kinetic data were collected simultaneously using an AMTI force plate (AMTI Corp., Watertown, MA). Positional data were collected at 200 Hz, with force data collected synchronously at 1000 Hz.

The Biodex System III power head was used for resistance to hip abduction during strength testing and the fatigue protocol (Biodex, Inc., Shirley, NY). The torque produced before and during the fatigue protocol was captured from the Biodex with custom software written in Labview 8.2 (National Instruments Corporation, Austin, TX).


The testing procedure was split into 2 d, 1 wk apart (Fig. 1). The first day was to familiarize the participant with the functional tasks to be collected and to document the effects of the fatigue protocol on force production of the hip abductors during the immediate postfatigue period. For this study, the fatigued state was defined as the inability of the participant to produce greater than 80% of the maximum measured hip abduction torque. This definition has been used previously to define the fatigued versus nonfatigued state when evaluating hamstring muscle function (34). In addition, several studies report a decrease in hip abduction strength on the side of injury when compared with control and uninjured lower extremities (10,12,23,27). In these studies, the difference between injured and noninjured hip abductors was approximately 20%. Thus, defining a decrease of 20% in force production as "fatigued" was supported in a previous study and was thought to be functionally significant. It is recognized that this fatigue protocol may have caused local muscle fatigue or may have resulted in central changes associated with fatigue such as decreased neural drive or a combination of both. However, because a decreased force-producing ability of the muscle was the parameter of interest, muscle weakness was defined in this study as the inability of the muscle to produce at least 80% of maximal peak torque during an isometric contraction. It was determined through pilot testing that motion analysis data collection after the fatigue protocol could be accomplished in 2 min or less. Thus, an additional inclusion criterion was that a participant's hip abductor force production had to remain below 80% for 2 min after cessation of the practice day fatigue protocol. If the participant's hip strength returned too quickly on the practice day, they were disqualified from the study.

Participant flow diagram.

Fatigue protocol.

The dominant stance leg was used as the test leg throughout the study (19 left, 1 right). This was identified by the participant as the opposite leg from the one they would use to kick a ball for distance. Participants were positioned side lying in front of the isokinetic dynamometer for the fatigue protocol as shown in Figure 2. The speed of the isokinetic resistance was set at 60°·s−1 during concentric hip abduction and at 300°·s−1 during the adduction phase. The 300°·s−1 speed during adduction was chosen to provide no resistance to adduction. Participants were instructed to not allow their leg to free fall and to lower their leg at approximately the same speed as they raised it. Thus, the participant was forced to use the hip abductor muscles to eccentrically lower the leg into adduction. Instructions were standardized to raise the leg into abduction as hard as possible, then lower the leg into adduction taking approximately 1 s to descend. As the muscle fatigued in the concentric portion first, when participants could no longer raise the leg unassisted, assistance was given during raising, and the participant was responsible for lowering the leg slowly. When the descent could not be controlled during the 1-s period, the muscle was also fatigued in the eccentric contraction. Participants performed an average of 105 repetitions (±27, range 67-147 reps) during this fatigue protocol.

Positioning on Biodex for fatigue protocol.

On day 1 of the study, participants performed the fatigue protocol as explained previously. When final fatigue was reached, they remained attached to the Biodex, and hip abductor isometric force production was retested at 30 s, 1 min, 2 min, and 3 min after the cessation of the fatigue protocol. If participants demonstrated the appropriate fatigue time course on the practice day (remained <80% of the peak torque for at least 2 min), they were taught the study tasks and were scheduled for data collection not sooner than 1 wk in the future. They were asked not to alter their normal activity pattern or change their workout routines during the interim period but to avoid aggressive lower extremity exercise.


During data collection on the second testing day, kinematic and kinetic data were collected during three tasks designed and chosen to represent the movements and forces associated with ACL injuries and PFPS. The two most common mechanisms for ACL injury involve a sudden deceleration on the involved lower extremity or a similar activity followed by a cut laterally (2). Thus, tasks that presented kinematically and kinetically similar to cutting, jumping, and running movements were chosen. The timing of data collection after the fatigue protocol necessitated that tasks be collected expediently, with minimal repeating of faulty trials. A running or cutting task in the laboratory would have likely required multiple trials to achieve correct foot placement on the force plate without an alteration of normal gait mechanics. Through pilot testing, three tasks that allowed both expedient data collection after fatigue and presented movement characteristics similar to planting, cutting, and running activities were identified. These tasks all started on a platform (MF Athletic, Cranston, RI) equal in height to the participant's vertical countermovement jump height. Their jump height was determined on day 1 of the study by having the participant perform a maximum jump on the force plate. Vertical jump height was then calculated using the impulse-momentum relationship. This platform height was chosen for the task starting position as representative of the highest position vertically from which participants would have to land during athletic competition, such as when landing from a rebound in basketball. This has been used previously in published studies (16). The platform was set away from the center of the force plate a distance equal to each participant's maximal stride length on level ground. Thus, when participants landed on the force plate during data collection, they had a standardized, controlled forward and downward velocity. Participants started all tasks standing on the platform on the nontest leg. They leaped forward, landing on the force plate with the test leg, and as fast as they could performed one of the three tasks. The first task was a sidestep cutting maneuver (Fig. 3) called the "cut." After landing, the participant quickly cut to the opposite side of the test leg, taking a large step at a 90° angle to the direction of approach, as if they were "cutting in basketball." This task was chosen as representative of an ACL injury mechanism involving direction change. During the second task, after landing, the participant quickly performed a maximal vertical jump off the test leg, called the "jump." This task involved a straight plane deceleration, another noted ACL injury mechanism. For the third task, after landing, the participant continued running forward in the same straight plane that they approached the force plate from, called the "run." In pilot testing, this task kinetically seemed similar to running without the uncertainty of the test leg falling on the force plate during an uninterrupted stride. Starting each task above ground height and away from the force plate increased the vertical and horizontal velocity of the participant in a consistent manner, replicating a running start at a constant speed for each task. The chosen tasks were also of sufficient rigor that alterations due to changes in maximal force production would likely be detectable. Participants were allowed to practice each task numerous times until comfortable with the performance of the tasks and until the investigator was comfortable with the participant's technique.

Study tasks. Participants started on the elevated box, leaped forward landing unilaterally on the test leg (left shown), then performed either a sidestep cut (shown), a straight vertical jump, or continued to run straight forward from that point.

Data collection.

Participants had 25 reflective markers placed on their body for the collection of kinematic data. Participants wore standardized Saucony Jazz running shoes from the laboratory, which have a neutral midsole to prevent variance due to footwear. After collection of a standing trial, both iliac crest markers, both greater trochanter markers, the medial and lateral knee markers, the medial and lateral malleolus markers, and the first and fifth metatarsal head markers were removed, leaving both anterior superior iliac spine (ASIS) and posterior superior iliac spine (PSIS) markers, and all three clusters (thigh, shank, heel) for segment identification.

The tasks were reviewed and participants allowed to practice one to two repetitions of each to feel comfortable with task performance. Three trials of each task were then collected before the fatigue protocol. Sufficient rest was given to prevent fatigue from affecting each subsequent trial. After the prefatigue data collection, the fatigue protocol was performed. Care was taken to ensure that the markers remained in place during performance of the fatigue protocol. The resistance arm of the Biodex was carefully aligned with the lateral knee between the thigh and the shank clusters so that no contact occurred throughout the fatigue protocol (Fig. 2). All markers remained well clear of the resistance arm. Hip abductor torque during the entire fatigue protocol was recorded. When participants could no longer control the descent of their leg in the eccentric phase, the fatigue protocol was terminated. A timer was immediately started and the dynamometer was again locked at 0° of abduction to record the torque from one final isometric contraction. These recordings were used to compare the test day fatigue data with the practice day fatigue timeline to ensure that the participants were truly fatigued enough to affect hip abductor performance for 2 min.

The participants then quickly repeated the three movement tasks, one repetition of each sequentially to a maximum of three trials during the 2-min period after the fatigue protocol. The order in which tasks were performed was consistent from before to after the fatigue protocol for each participant but was counterbalanced (run, cut, jump) between participants throughout the study.

Data analysis.

Three-dimensional coordinate data were processed using Visual 3D software (C-Motion, Inc., Germantown, MD). The kinematic and kinetic data were filtered using a fourth-order low-pass Butterworth filter with a cutoff frequency of 18 Hz (30). Hip, knee, and ankle joint angles were calculated using a joint coordinate system approach (15). Joint centers for the knee and ankle were defined as the midpoint between the medial and lateral joint markers. The hip joint center was estimated at 25% of the horizontal distance between the greater trochanters from the test side trochanter marker (37). An inverse dynamics approach (3,24) was used to derive the joint kinetic data of the hip, knee, and ankle from the ground reaction force (GRF) and kinematic data. All moments reported are internal net joint moments and are reported in the distal segment. For clarity of presentation, during processing, left leg data were inverted in the frontal plane so that numeric results are all presented from the perspective of a right lower extremity. For all three tasks, the WA phase was defined as the period from when the vertical GRF (v-GRF) exceeds 30 N to the first trough in the v-GRF (1). Given that most ACL injuries occur within the first 20% of stance phase (3) and that the peak GRF for each task was within this phase for all three tasks, analysis was confined to just the WA phase. Dependent variables extracted for analysis were the frontal plane hip and knee angle at initial ground contact (IC), frontal plane hip and knee ROM during WA (defined as the excursion of the hip and knee from IC to maximum joint angle − ROM), and the mean frontal plane hip and knee moment during WA.

Statistical analysis.

Statistical analyses were performed using SPSS Version 16.0 (SPSS, Inc., Chicago, IL). Paired t-tests were performed between prefatigue and postfatigue maximal hip abductor peak torque for both practice and test days. In addition, the percentage of strength deficit was calculated by dividing the immediate postfatigue torque by the prefatigue torque multiplied by 100, and a paired t-test was performed to compare the strength deficit between days.

Separate repeated-measures ANOVA were performed to identify significant differences in the three dependent variables across the three tasks. The independent variables were fatigue with two levels (before and after fatigue) and task with three levels (cut, jump, and run). The dependent variables were the hip and knee frontal plane angles at IC, hip and knee joint frontal plane ROM during WA, and mean hip and knee frontal plane moments during WA. The significance level for each test was set a priori at P < 0.05.


Twenty-two participants underwent the orientation day fatigue protocol testing. One participant did not demonstrate the necessary force decrements in hip force production after the fatigue protocol and was disqualified after the practice day. Another participant opted not to continue after the practice day, leaving 20 participants with test day data for analysis.

Hip fatigue.

Hip abductor force production was significantly decreased after the fatigue protocol on both the orientation day (pre = 115 ± 21 ft·lb, post = 73 ± 17 ft·lb, t(38) = 19.49, P < 0.001) and the test day (pre = 102 ± 20, post = 64 ± 17, t(38) = 13.58, P < 0.001; Fig. 4). Although peak torque was greater on the orientation day than on the test day, both before and after the fatigue protocol, the strength deficit from before to after the fatigue protocol was not significantly different between orientation and test days (orientation day = 62% ± 7%, test day = 63% ± 11%, t(38) = −0.068, P = 0.946), indicating that the level of fatigue reached was the same between days (Fig. 4).

Torque produced by the hip abductors before the fatigue protocol, at cessation of the fatigue protocol, and 2 min after cessation of the fatigue protocol. Pilot testing demonstrated that the recovery profile on the test day at 2 min after cessation of the protocol mirrored the recovery timeline documented on the orientation day. *Significant difference (P < 0.05) from baseline (max) that day.

Joint dynamics.

The frontal plane knee angle (Fig. 5) and moment (Fig. 6) time series are presented, and individual task by variable results are found in Table 1. Analysis of the frontal plane hip and knee data revealed no significant fatigue × task interactions for any of the dependent variables. Main effects for fatigue were noted for all three frontal plane dependent variables at the hip and the knee. After the hip abductor fatigue protocol, the hip angle at IC was more abducted (difference of 1.6° ± 6.2°, P < 0.001), the hip ROM during the WA phase was greater from abduction to adduction (difference of 0.7° ± 2.2°, P = 0.006), and mean hip abductor moment during the WA phase decreased (difference of 4.2 ± 21.9 N·m, P = 0.016). The knee angle at IC was more adducted (difference of 0.5 ± 3.1 deg., P = 0.032), knee ROM during WA moved from a greater adducted position to a greater abducted position (1.4° ± 1.3°, P < 0.001), and the knee moment shifted toward a greater internal adductor moment during WA (difference of 7.4 ± 13.4 N·m, P < 0.001). This indicates that after the hip fatigue protocol, participants landed initially in a more varus knee position and shifted to a greater valgus knee position as the hip collapsed into more adduction, while the knee experienced a greater mean knee adductor moment throughout WA, regardless of task (Table 1).

rontal plane hip and knee angles for the cut, run, and jump over stance phases (normalized to 100%).Thick black line denotes the average end of the WA phase.
rontal plane hip and knee moments for the cut, run, and jump over stance phases (normalized to 100%).Thick black line denotes the average end of the WA phase.
Mean ± SD hip and knee frontal plane angle at IC, ROM during WA, and moment for cut, jump, and run tasks.

Main effects for task were also noted for each variable. There was a difference among the cut, jump, and run in all hip and knee dependent variables. However, hip abductor fatigue did not affect each task differently (Table 1).


The results of this study indicate that fatigue-induced hip weakness increased the motion experienced by the knee and shifted the moment at the knee in the frontal plane. The hypothesis that hip abductor fatigue would increase knee abduction ROM was supported. Knee internal adductor moment increased during the cut and the jump, and although the moment shifted in the same direction during the run, this shift resulted in a decrease in internal knee abductor moment. These differences were measured during WA, the period when most ACL injuries occur (2,35). In addition, the changes noted in knee angle and moment during the cut and the jump tasks occurred in the direction previously associated with an increased risk of ACL injury (17) and with knee injury mechanics (2,35,36), suggesting that hip abductor weakness may play a role in knee injury risk.

The hip abductor fatigue protocol had a small effect on all three knee dependent variables. After fatigue, the knee started in a more adducted position at IC, then moved to a more abducted position during WA. During the cut and the jump, this shift was consistent with the movement pattern defined in the literature as dynamic valgus of the knee (9). Correspondingly, the frontal plane knee joint moment during the cut and the jump shifted to a greater internal knee abductor moment after fatigue. The cut and jump tasks are representative of those movements noted during ACL injuries (2,35). Although the average changes noted are small in magnitude, all subjects individually demonstrated ROM changes in the same direction of knee abduction and all but one demonstrated an increased knee adductor moment after fatigue (Figs. 7 and 8). Withrow et al. (39) reported that a 2.49° increase in knee abduction increased ACL strain by 30% in vitro. Although there are substantial differences in study design and the participants/specimens used, there is an indication that small movements into knee abduction may be important to knee injury mechanics. Interestingly, although these changes occur in the same direction during all tasks, the changes during the run moved the knee toward a more neutral position or less adducted position, and the abductor moment decreased during this task.

hange in frontal plane knee ROM (postfatigue minus prefatigue) for individual participants across all tasks.Negative numbers indicate a participant demonstrating greater knee ROM, from a more adducted position to a more abducted position when the hip abductors were fatigued.
hange in frontal plane knee moment (postfatigue minus prefatigue) for individual participants across all tasks.Positive numbers indicate a greater internal knee adductor moment when the hip abductors were fatigued.

Similar task differences were noted for the frontal plane knee moment. During the cut and the jump, the frontal plane moment shifted to a greater internal knee adductor moment. The increases observed in the frontal plane knee adductor moment during these two tasks are in agreement with the results of the computer model used to evaluate frontal plane knee forces with changes in hip abductor stiffness (6), which demonstrated that a 50% decrease in hip abductor stiffness resulted in increases in the frontal plane knee moment in the direction of increased internal knee adductor moments in response to axial loading of the joint. This effect was most pronounced when the valgus angle of the knee was increased. However, in the current study, with before and after differences in this small range, we cannot rule out that measurement error during motion analysis from skin movement over osseous structures or a systematic shift in marker position during the fatigue protocol did not account for the differences noted. We did not collect a postfatigue standing trial to ensure that markers did not shift during testing. Every precaution was taken to prevent this occurrence, but we acknowledge the possibility that this type of error occurred.

One previous study reported that hip abductor fatigue increased knee abduction angle in the frontal plane at initial contact but not the maximum frontal plane angle (5). They used a combination of isometric open and closed chain hip abduction exercises to produce fatigue and measured frontal plane movement during a bilateral drop landing using electrogoniometers. The present study does not support these findings because the frontal plane angle at initial contact was more varus after fatigue and then collapsed into more valgus throughout WA. The contradictory findings are perhaps caused by the differences in the task, fatigue protocol, and data collection equipment. Generally, the force required from the hip abductors increases greatly during a unilateral task because of the increased torque applied to the hip by body weight. With a bilateral task, demands on the hip abductors are submaximal, and changes in hip abductor force production because of the fatigue may not be apparent because the demands on the muscle may not reach a critical level. The single-leg tasks in the present study were thought to be demanding on the hip abductors, requiring both ballistic eccentric control of the body during a forceful landing and immediate concentric activity to propel the individual in the desired postlanding direction. Differences in measurement accuracy of the three-dimensional camera system compared with the electrogoniometers, particularly in the frontal plane, could also account for the conflicting results. Finally, although the combination of open and closed chain exercises used previously may have greater practical or functional application, our use of repeated concentric and eccentric open chain hip abduction as the fatiguing activity more likely isolated the hip abductor muscle group and is likely to have produced greater fatigue in that specific group.

Of note is that absolute angles are reported in this study. This is a representation of an individual's anatomy on the basis of surface markers. There may be small errors in marker placement for the identification of joint centers, which is a limitation to this technique of capturing human movement. The anatomic angles identified in the standing calibration were not cross checked with imaging studies. However, the markers were not removed between the prefatigue and postfatigue trials, so error due to marker placement is minimized. The results of this study also document the change from before to after fatigue regardless of the initial anatomic position of the knee. Whereas that may or may not increase the risk of knee injury to those individuals who are anatomically adducted initially, the results demonstrate that hip fatigue-induced weakness would move someone who is initially in a valgus or neutral position to a further valgus position at the knee and, therefore, may increase the risk of injury for those marginal individuals.

Leetun et al. (27) reported that hip abductor strength was different between collegiate athletes who experienced lower extremity injuries and those who did not in a prospective study on the effect of core strength on injury rates, although hip external rotator strength difference was the only predictor of lower extremity injury. Many other cross-sectional studies have reported impaired strength or function of the hip abductors of injured individuals or of the injured lower extremity (4,10,23,33). The results of the present study lend insight into the possible mechanical effects by which hip abductor weakness increases knee injury risk.

Interestingly, hip abductor fatigue in this study created a very small change toward an adducted (varus) knee position (0.5° change) and an abducted position of the hip (1.6° change) at IC, possibly indicating an anticipatory response to the acute hip abductor weakness. Although this positional change is very small in magnitude, the results were consistent enough to reach statistical significance. The sensation of hip abductor fatigue may have caused participants to alter their movement patterns in a compensatory effort to maximize force absorption time and ROM. Previous studies have identified kinematic and kinetic differences between anticipated and unanticipated maneuvers (20). Had unanticipated tasks been tested, participants may not have been able to compensate for hip abductor insufficiency before task initiation, and the results may have been altered. We also cannot rule out that these small positional changes are not from a systematic shift in marker position during the fatigue protocol and may in fact represent artifact from such an occurrence.

In addition to small alterations in frontal plane knee joint mechanics, alterations in hip joint variables were also noted after fatigue. Hip abductor moment was expected to decrease after fatigue because the muscle could not produce the same force as in the unfatigued state, and that was indeed observed. The hip joint ROM in the frontal plane was increased after fatigue due largely to an increase in the IC abduction angle. Hip abduction angle at IC increased an average of 1.6° (Table 1), whereas hip ROM through WA increased by 0.7°.

Stefanyshyn et al. (38) reported that, prospectively, runners who demonstrated a larger internal knee abductor impulse were more likely to develop PFPS during the course of a season. Participants in the current study did not demonstrate changes in knee moment in the direction described by Stefanyshyn et al. The knee moment changes in this study after fatigue were in the direction of greater knee adductor moments, which is the opposite direction of the differences described by Stefanyshyn et al. (38). The run task evaluated in the present study was chosen because it did not introduce the uncertainty of whether the foot would fall naturally on the force plate during the time-limited postfatigue data collection. Pilot testing demonstrated that this task seemed kinematically and kinetically similar to actual running. Prefatigue values for knee angle and moment seem consistent with the kinematic and kinetic patterns reported in previous running studies (29,38), the major difference being an increased v-GRF component during WA due to the height that the current participants started the task from. However, it is likely that the run task used in the present study did not truly represent running and that our results are not indicative of running as an activity. The length of the stride and the vertical acceleration while coming off the platform may have created more focus on the individual stride landing than during overground running, thus altering knee behavior. It is also possible that the mechanism behind the development of PFPS in runners in the study of Stefanyshyn et al. is different from the mechanism by which hip weakness potentially influences the development of PFPS. Stefanyshyn et al. did not report hip strength measures for their participants, so the role of hip strength as it relates to the development of PFPS will require further study.

In this study, force decrements associated with fatigue were used to represent force decrements that occur with muscle weakness of the hip abductors. It should be noted that the kinematic and kinetic changes noted with muscle fatigue might not be the same as those seen with muscle weakness. Muscle weakness occurs during an extended period, allowing for the development of compensation patterns in response to the weakness. Those long-term changes cannot be detected by this study design. In addition, decreases in the force-producing capacity of the muscle produced by fatigue are accompanied by other neuromuscular changes as well (14). Fatigue was used to alter the force-producing capability of the hip abductor muscles in an attempt to draw conclusions about the effect of hip weakness on lower extremity motion. Although muscle force was affected by the fatigue protocol, it cannot be elucidated whether the findings in the present study represent changes due to weakness, due to alterations in neuromuscular control of the fatigued muscle, or due to some other component of fatigue. In addition, we did not test the response of the fatigued hip abductor muscle to stretch-shortening type of activities. The response to a stretch-shortening type of activity, like the ballistic tasks used in this study, may be different for a fatigued muscle from a chronically weak muscle. This is a limitation of this study design. Hip abductor fatigue also does not occur in isolation during athletic activity, and this study sought to specifically and selectively fatigue just the hip abductor muscle group. This limits the generalizability of these results during athletic activity.

The tasks in this study were taught to participants on the orientation day only after they demonstrated satisfactory performance of the fatigue protocol. A recovery period was allowed; nevertheless, participants thus learned the tasks although their hip abductors were not fully recovered. This may have ultimately affected the initial performance of the tasks on day 2, the test day, and thus affected the results of this study.

Consideration must be given to why the differences in postfatigue knee variables were not greater and if in fact the small magnitude of the changes means that hip weakness is indeed not a major factor in knee injuries. The small changes noted here are not of sufficient magnitude to be clinically meaningful and could be the result of undetected systematic artifact. It could be concluded from these results that hip abductor fatigue does not affect knee mechanics in a meaningful way. However, neuromuscular control of voluntary movement is a complex topic, and the fact that statistical changes were noted at the knee while participants knew the task and the direction of movement is of interest. Further study using these techniques with unanticipated tasks or directions of movement would be a logical progression of this work.

In summary, the purpose of this study was to identify the acute effects of hip abductor weakness, simulated by an isolated fatigue protocol, on knee mechanics during strenuous unilateral athletic activities. Small changes were noted in the frontal plane dependent variables, the plane of the hip abductors main force-producing ability. While in the fatigued state, during the cut and the jump tasks, participants demonstrated greater knee abduction ROM and a shift in the knee moment to a greater internal adductor moment across tasks during WA. The postfatigue changes at the knee during these two tasks were in the direction identified in the literature as increasing knee injury risk, particularly for ACL injury. Further evaluation of the effect of hip abductor weakness on knee mechanics is warranted.

The authors thank the Great Lakes Athletic Trainer's Association for funding of this research, the participants for their participation in this study, Dr. Susan Cashin for her assistance with the statistical design and analysis of the data, and Dr. Barbara Hart for her assistance with the development and design of this study. The results of the present study do not constitute endorsement by American College of Sports Medicine.


1. Bobbert MF, van Zandwijk JP. Dynamics of force and muscle stimulation in human vertical jumping. Med Sci Sports Exerc. 1999;31(2):303-10.
2. Boden BP, Dean GS, Feagin JA Jr, Garrett WE Jr. Mechanisms of anterior cruciate ligament injury. Orthopedics. 2000;23(6):573-8.
3. Bresler B, Frankel JP. The forces and moments in the leg during level walking. Trans Am Soc Mech Eng. 1950;72:27-36.
4. Brindle TJ, Mattacola C, McCrory J. Electromyographic changes in the gluteus medius during stair ascent and descent in subjects with anterior knee pain. Knee Surg Sports Traumatol Arthrosc. 2003;11(4):244-51.
5. Carcia C, Eggen J, Shultz S. Hip-abductor fatigue, frontal-plane landing angle, and excursion during a drop jump. J Sport Rehabil. 2005;14(4):321-31.
6. Chaudhari AM, Andriacchi TP. The mechanical consequences of dynamic frontal plane limb alignment for non-contact ACL injury. J Biomech. 2006;39(2):330-8.
7. Christoforakis JJ, Strachan RK. Internal derangements of the knee associated with patellofemoral joint degeneration. Knee Surg Sports Traumatol Arthrosc. 2005;13(7):581-4.
8. Csintalan RP, Ehsan A, McGarry MH, Fithian DF, Lee TQ. Biomechanical and anatomical effects of an external rotational torque applied to the knee. Am J Sports Med. 2006;34(10):1623-29.
9. Earl JE, Hertel J, Denegar CR. Patterns of dynamic malalignment, muscle activation, joint motion, and patellofemoral-pain syndrome. J Sport Rehabil. 2005;14(3):215-33.
10. Fredericson M, Cookingham CL, Chaudhari AM, Dowdell BC, Oestreicher N, Sahrmann SA. Hip abductor weakness in distance runners with iliotibial band syndrome. Clin J Sport Med. 2000;10(3):169-75.
11. Freedman KB, Glasgow MT, Glasgow SG, Bernstein J. Anterior cruciate ligament injury and reconstruction among university students. Clin Orthop. 1998;(356):208-12.
12. Friel K, McLean N, Myers C, Caceres M. Ipsilateral hip abductor weakness after inversion ankle sprain. J Athletic Train. 2006;41(1):74-8.
13. Gabriel MT, Wong EK, Woo SL, Yagi M, Debski RE. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J Orthop Res. 2004;22(1):85-9.
14. Gribble PA, Hertel J. Effect of lower-extremity muscle fatigue on postural control. Arch Phys Med Rehabil. 2004;85(4):589-92.
15. Grood ES, Suntay WJ. A joint coordinate system for the clinical description of three-dimensional motions: application to the knee. J Biomech Eng. 1983;105(2):136-44.
16. Hass CJ, Schick EA, Tillman MD, Chow JW, Brunt D, Cauraugh JH. Knee biomechanics during landings: comparison of pre- and post-pubescent females. Med Sci Sports Exerc. 2005;37(1):100-7.
17. Hewett TE, Myer GD, Ford KR, et al. Biomechanical measures of neuromuscular control and valgus loading of the knee predict anterior cruciate ligament injury risk in female athletes: a prospective study. Am J Sports Med. 2005;33(4):492-501.
18. Hewett TE, Stroupe AL, Nance TA, Noyes FR. Plyometric training in female athletes. Decreased impact forces and increased hamstring torques. Am J Sports Med. 1996;24(6):765-73.
19. Hollis JM, Takai S, Adams DJ, Horibe S, Woo SL. The effects of knee motion and external loading on the length of the anterior cruciate ligament (ACL): a kinematic study. J Biomech Eng. 1991;113(2):208-14.
20. Houck JR, Duncan A, De Haven KE. Comparison of frontal plane trunk kinematics and hip and knee moments during anticipated and unanticipated walking and side step cutting tasks. Gait Posture. 2006;24(3):314-22.
21. Huston LJ, Greenfield ML, Wojtys EM. Anterior cruciate ligament injuries in the female athlete. Potential risk factors. Clin Orthop. 2000;(372):50-63.
22. Ireland ML. The female ACL: why is it more prone to injury? Orthop Clin North Am. 2002;33(4):637-51.
23. Ireland ML, Willson JD, Ballantyne BT, Davis IM. Hip strength in females with and without patellofemoral pain. J Orthop Sports Phys Ther. 2003;33(11):671-6.
24. Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res. 1990;8(3):383-92.
25. Kao JT, Giangarra CE, Singer G, Martin S. A comparison of outpatient and inpatient anterior cruciate ligament reconstruction surgery. Arthroscopy. 1995;11(2):151-6.
26. Lee TQ, Morris G, Csintalan RP. The influence of tibial and femoral rotation on patellofemoral contact area and pressure. J Orthop Sports Phys Ther. 2003;33(11):686-93.
27. Leetun DT, Ireland ML, Willson JD, Ballantyne BT, Davis IM. Core stability measures as risk factors for lower extremity injury in athletes. Med Sci Sports Exerc. 2004;36(6):926-34.
28. Maletius W, Messner K. Eighteen- to twenty-four-year follow-up after complete rupture of the anterior cruciate ligament. Am J Sports Med. 1999;27(6):711-7.
29. McClay I, Manal K. Three-dimensional kinetic analysis of running: significance of secondary planes of motion. Med Sci Sports Exerc. 1999;31(11):1629-37.
30. McLean SG, Huang X, van den Bogert AJ. Association between lower extremity posture at contact and peak knee valgus moment during sidestepping: implications for ACL injury. Clin Biomech. 2005;20(8):863-70.
31. McLeish RD, Charnley J. Abduction forces in the one-legged stance. J Biomech. 1970;3(2):191-209.
32. Miyasaka KC, Daniel DM, Stone ML, Hirshman P. The incidence of knee ligament injuries in the general population. Am J Knee Surg. 1991;4(1):3-8.
33. Niemuth PE, Johnson RJ, Myers MJ, Thieman TJ. Hip muscle weakness and overuse injuries in recreational runners. Clin J Sport Med. 2005;15(1):14-21.
34. Nyland JA, Caborn DNM, Shapiro R, Johnson DL. Crossover cutting during hamstring fatigue produces transverse plane knee control deficits. J Athletic Train. 1999;34(2):137-43.
35. Olsen OE, Myklebust G, Engebretsen L, Bahr R. Injury mechanisms for anterior cruciate ligament injuries in team handball: a systematic video analysis. Am J Sports Med. 2004;32(4):1002-12.
36. Powers CM. The influence of altered lower-extremity kinematics on patellofemoral joint dysfunction: a theoretical perspective. J Orthop Sports Phys Ther. 2003;33(11):639-46.
37. Robertson DG, Caldwell GE, Hamill J, Kamen J, Whittlesey SN. Research Methods in Biomechanics. Champaign (IL): Human Kinetics; 2004. 309 p.
38. Stefanyshyn DJ, Stergiou P, Lun VMY, Meeuwisse WH, Worobets JT. Knee angular impulse as a predictor of patellofemoral pain in runners. Am J Sports Med. 2006;34(11):1844-51.
39. Withrow TJ, Huston LJ, Wojtys EM, Ashton-Miller JA. The effect of an impulsive knee valgus moment on in vitro relative ACL strain during a simulated jump landing. Clin Biomech. 2006;21(9):977-83.
40. Witvrouw E, Lysens R, Bellemans J, Cambier D, Vanderstraeten G. Intrinsic risk factors for the development of anterior knee pain in an athletic population. A two-year prospective study. Am J Sports Med. 2000;28(4):480-9.


©2010The American College of Sports Medicine