Hamstring musculotendon dynamics during stance and swing phases of high-speed running

Abstract

Introduction: Hamstring strain injuries are common in sports that involve high-speed running. It remains uncertain whether the hamstrings are susceptible to injury during late swing phase, when the hamstrings are active and lengthening, or during stance, when contact loads are present. In this study, we used forward dynamic simulations to compare hamstring musculotendon stretch, loading, and work done during stance and swing phases of high-speed running.

Methods: Whole-body kinematics, EMG activities, and ground reactions were collected as 12 subjects ran on an instrumented treadmill at speeds ranging from 80% to 100% of maximum (avg max speed = 7.8 m·s(-1)). Subject-specific simulations were then created using a whole-body musculoskeletal model that included 52 Hill-type musculotendon units acting about the hip and the knee. A computed muscle control algorithm was used to determine muscle excitation patterns that drove the limb to track measured hip and knee sagittal plane kinematics, with measured ground reactions applied to the limb.

Results: The hamstrings lengthened under load from 50% to 90% of the gait cycle (swing) and then shortened under load from late swing through stance. Although peak hamstring stretch was invariant with speed, lateral hamstring (biceps femoris) loading increased significantly with speed and was greater during swing than stance at the fastest speed. The biarticular hamstrings performed negative work on the system only during swing phase, with the amount of negative work increased significantly with speed.

Conclusion: We concluded that the large inertial loads during high-speed running appear to make the hamstrings most susceptible to injury during swing phase. This information is relevant for scientifically establishing muscle injury prevention and rehabilitation programs.

Figure 1

A graphical depiction of how the simulations were generated. (a) Kinematic and kinetic data were first collected as each subject sprinted on a treadmill. (b) A computed muscle control algorithm (31) was used to generate the excitation patterns that drove the limb to closely track measured hip and knee kinematics. (c) Excitations were inputs into Hill-type models of musculotendon dynamics, which produce force that drive sagittal plane hip and knee motion. All other degrees of freedom were prescribed to follow measured trajectories, while ground reaction forces and moments were applied directly on the lower limbs. (d) Simulations were then generated, from which hamstring force, negative and positive work could then be determined.

Figure 2

At maximal sprinting speed (100%), the timing of simulated muscle excitations (solid lines, average over all subjects) and measured electromyographic (EMG) activities (shaded curves) are shown to be relatively consistent. EMG activities are the mean (± 1 s.d.) rectified, low-pass filtered activities over all subjects. In the model, the medial hamstrings (semitendinosus and semimembranosus) and iliopsoas (iliacus and psoas) are represented as two individual muscle line segments. Three line segments in the model were used to represent the broad attachments of gluteus maximus.

Figure 3

The average musculotendon mechanics of the hamstring muscles over a simulated sprinting gait cycle for a representative subject. The length excursions (Δℓ ^MT) of the hamstring musculotendons are relatively consistent across running speeds, with the lengthening phase constrained to swing phase. Peak swing phase musculotendon forces increased with speed for each of the hamstring muscles while stance phase peak forces remain invariant with speed. The hamstrings are stretching and do negative work (i.e. integral of negative power) on the system from 50 to 90% of the gait cycle, and then shorten and do positive work at the end of swing through stance phase.

Figure 4

Peak biceps femoris loading during swing phase significantly (p<0.001) increases with speed, and significantly (p<0.003) exceeds the loading incurred during stance at the fastest speed. Both net negative work and positive work increase as running speed increases, with net negative work increasing at a faster rate with speed than net positive work. The net work values shown are the sum of all three biarticular hamstring work (negative and positive).