Hip-Extensor Strength, Trunk Posture, and Use of the Knee-Extensor Muscles During Running

A total of 40 recreational runners (20 men, 20 women) participated (Table 1). They ran at least 8.05 km per week and were natural heel strikers, which was verified using sagittal-plane images obtained from high-speed video with a sampling rate of 125 Hz. Volunteers were excluded from participation if they reported any of the following: (1) lower extremity or low back pain at the time of the study, (2) a history of lower extremity or low back surgery, and (3) a lower extremity or low back pathologic condition that caused pain or discomfort during running in the 6 months before the study. All participants provided written informed consent, and the study was approved by the University of Southern California Health Sciences Institutional Review Board.

Hip-extensor strength was assessed using a motor-driven dynamometer (Cybex with HUMAX NORM; Computer Sports Medicine Inc, Stoughton, MA). The dynamometer provided force values in newton-meters, with a precision of 0.02% (full scale). The sampling frequency was 100 Hz. Three-dimensional trunk and lower extremity kinematics were collected using an 11-camera motion-capture system (Qualisys, Gothenburg, Sweden) at a sampling rate of 250 Hz. Ground reaction forces were obtained at a sampling rate of 1500 Hz using a single force plate (AMTI, Newton, MA).

Participants wore shorts, tank tops, and their personal running shoes during the evaluation. We obtained data from the dominant leg, which was defined as the lower extremity that the participant preferred to use when kicking a ball.

To assess isometric hip-extensor strength, participants were positioned prone on the dynamometer testing table with the nondominant leg on the ground (Figure 1). The dominant leg was positioned with the hip and knee at 60° and 90° of flexion, respectively.^25,27  We aligned the axis of the dynamometer with the greater trochanter of the dominant leg. The lower end of the resistance pad was positioned just proximal to the lateral knee-joint line and secured to the distal thigh with straps. We instructed participants to push with maximal effort against the resistance pad. Oral encouragement was given throughout testing. Three trials of 5-second maximal voluntary isometric contractions were obtained. We provided a 40-second break between trials to minimize muscle fatigue.²⁸  Participants performed 2 practice trials before the test to become familiar with the task.

After the strength measurement, we placed 21 reflective 14-mm spherical anatomical markers on the following bony landmarks: ends of the second toes, first and fifth metatarsal heads, medial and lateral malleoli, medial and lateral epicondyles of the femurs, greater trochanters, iliac crests, L5–S1 junction, and acromioclavicular joints. In addition, tracking marker clusters mounted on semirigid plastic plates were placed on the lateral surfaces of the thighs, shanks, and heel counters of the shoes. We obtained a standing calibration trial to define the segmental coordinate systems and joint axes. After the calibration trial, the anatomical markers were removed except for those at the iliac crests, L5–S1 junction, and acromioclavicular joints. The tracking markers remained on participants throughout the entire data-collection session.

Next, participants were instructed to run at a controlled speed of 3.4 m/s along a 14-m runway. We used a standardized velocity because running speed can influence hip and knee kinetics.^7,8,29  Two photoelectric sensors were placed 6 m apart in the middle of the runway to delineate the data-collection area. Running speed was calculated within the 6-m data-collection area. Practice trials were permitted to allow participants to become familiar with the running speed. Five successful running trials were obtained. A trial was considered successful if the foot of the dominant leg fell within the borders of the force plate and the running speed was ±5% of the target velocity.

During the strength test, torque production of the hip-extensor muscles was quantified using MATLAB software (The MathWorks Inc, Natick, MA). Specifically, the maximum 1-second average obtained during the strength measurement was exported for statistical analysis. Kinematic and kinetic data were processed and analyzed using Visual3D software (C-Motion, Germantown, MD). Data for the marker trajectories were low-pass filtered at 12 Hz using a fourth-order Butterworth filter. We defined the trunk segment by markers placed on bilateral iliac crests and acromioclavicular joints.³⁰  We modeled the pelvis and trunk segments as cylinders and the lower extremity segments as frusta of cones. The local orthogonal coordinate systems of the trunk, pelvis, thigh, shank, and foot segments were derived from the standing calibration trial.

Joint kinematics were calculated using the Cardan rotation sequence in the order of flexion-extension, abduction-adduction, and internal-external rotation.³¹  We calculated the trunk angle as the motion of the trunk segment relative to the global coordinate system (global vertical axis). Lower extremity kinematics were calculated as the motion of the distal segment relative to the proximal reference. Net joint moments were computed using inverse-dynamics equations. Kinetic data were expressed as internal moments and normalized to body mass.

We calculated net joint power as the scalar product of angular velocity and net joint moment. The positive and negative values of the joint power were used to identify phases of energy generation and absorption, respectively.^3,32,33  We calculated energy generation and absorption performed by the hip and knee extensors by integrating the respective power-time curves during the stance phase of running. For example, energy absorbed by the hip extensors was computed as the integral of negative power with respect to the time when the hip-extensor moment was positive. The absolute values of energy absorption and generation were summed to provide an estimation of the total work performed by the hip and knee extensors.

Mean trunk-flexion angle, hip-extensor work, and knee-extensor work during the stance phase were exported for statistical analysis. We defined the stance phase as when the vertical ground reaction force exceeded 30 N. All variables were calculated for each stride, and the average values of 5 strides were used for statistical analysis.

Pearson product moment correlations were used to examine the relationships among hip-extensor strength, mean sagittal-plane trunk-flexion angle, hip-extensor work, and knee-extensor work. To control for the influence of sex, partial correlations were performed to examine the aforementioned relationships. We set the α level at .05. All statistical analyses were performed using SPSS (version 22.0; IBM Corporation, Armonk, NY).

Results of the Pearson product moment correlations and partial correlations are presented in Table 2. A positive correlation between hip-extensor strength and trunk-flexion angle was observed (r = 0.55, P < .001; Figure 2). Moreover, hip-extensor strength exhibited a positive correlation with hip-extensor work (r = 0.46, P = .003; Figure 3) and an inverse correlation with knee-extensor work (r = −0.39, P = .01; Figure 4). After adjusting for sex, hip-extensor strength remained correlated with trunk-flexion angle, hip-extensor work, and knee-extensor work (Table 2).

We sought to elucidate the influence of hip-extensor strength on sagittal-plane trunk posture and hip and knee energy absorption and generation during running. As hypothesized, hip-extensor strength was correlated with trunk posture and hip- and knee-extensor work during running. In total, hip-extensor strength explained 30.5% of the variance in the mean sagittal-plane trunk posture and 21.2% and 15.2% of the variance in the hip- and knee-extensor work, respectively.

Our finding that hip-extensor strength was positively correlated with trunk-flexion angle is in contrast to that of Ford et al,²⁶  who reported no correlation between hip-extensor strength and sagittal-plane trunk motion. The discrepancy between our study and that of Ford at el²⁶  may be due to the differences in methods. First, Ford et al²⁶  quantified trunk motion using range of motion rather than the actual angles that we reported. Whereas range of motion of the trunk may indicate trunk stability, trunk position has been found to be related to lower extremity moments, work, muscle activation, and joint stress^3,4,19,30,34  and, thus, may be a more important variable in lower extremity mechanics. Second, Ford et al²⁶  evaluated isokinetic hip-extensor strength; we measured isometric strength. Whereas isokinetic strength may better reflect muscle performance during dynamic motion, we used isometric strength testing to obtain a general measure of muscle performance that could be reproduced more readily in a clinical setting without a dynamometer or with inexpensive equipment.

We also noted that individuals with weaker hip extensors demonstrated less hip-extensor and more knee-extensor work during running. This is consistent with the findings of Stearns et al,²⁵  who reported that individuals with diminished hip-extensor strength relative to knee-extensor strength exhibited higher knee-extensor moments relative to hip-extensor moments during a drop-jump task. Taken together, our observations and those of Stearns et al²⁵  suggest that diminished hip-extensor strength may contribute to an overreliance on the knee extensors during dynamic activities. An overreliance on the knee extensors indicates higher mechanical loads on the tibiofemoral and patellofemoral joints. For example, increased quadriceps force has been associated with increased patellofemoral-joint stress and patellofemoral pain^4,35–37and the presence and progression of patellofemoral-joint cartilage lesions.^38,39  Our observations may partially explain the results from previous prospective⁴⁰  and retrospective⁴¹  studies in which hip-muscle weakness was found to be associated with the development of various knee injuries.

The findings of our study suggest that strength training of the hip extensors will likely influence the mechanical demands on the hip and knee extensors during running. Further research is warranted to evaluate the effect of hip-extensor strength training on altering trunk posture and reducing knee injuries in runners. However, less than 31% of the variance in trunk posture and hip- and knee-extensor work was explained by hip-extensor strength.

Several limitations need to be considered when interpreting our results. First, given the cross-sectional design of the study, causal relationships cannot be drawn between hip-extensor strength and trunk and lower extremity biomechanics during running. For example, diminished hip-extensor strength may lead to or could result from a more upright trunk posture during running. Second, hip-extensor strength was quantified isometrically. Measures of muscle endurance or isokinetic muscle-performance testing may have yielded stronger correlations among trunk posture and lower extremity biomechanics. Third, we recruited only pain-free volunteers. Given that pain may influence running biomechanics, caution should be taken when generalizing the results to symptomatic runners. Fourth, participants were all heel strikers and ran at a controlled speed. It is unclear whether the observed correlation would hold true for forefoot and midfoot strikers or different running speeds. Fifth, we did not standardize running shoes because we wanted to ensure that the participants would run with their most natural form. However, the type of running shoe could have influenced knee and hip work.

Runners with diminished hip-extensor strength exhibited a more upright trunk posture, less work performed by the hip extensors, and more work performed by the knee extensors. In contrast, runners with greater hip-extensor strength exhibited a more forward-leaning trunk, more work performed by the hip extensors, and less work performed by the knee extensors. Our findings suggest that runners with hip-extensor weakness used a more upright trunk posture during running to minimize the demand on the hip extensors. In turn, this strategy appeared to lead to an overreliance on the knee extensors and may contribute to overuse running injuries at the knee.