Positive Work Contribution Shifts from Distal to Proximal Joints during a Prolonged Run

Purpose To investigate the joint-specific contributions to the total lower-extremity joint work during a prolonged fatiguing run. Methods Recreational long-distance runners (n = 13) and competitive long-distance runners (n = 12) performed a 10-km treadmill run with near-maximal effort. A three-dimensional motion capture system synchronized with a force-instrumented treadmill was used to calculate joint kinetics and kinematics of the lower extremity in the sagittal plane during the stance phase at 13 distance points over the 10-km run. Results A significant (P < 0.05) decrease of positive ankle joint work as well as an increase of positive knee and hip joint work was found. These findings were associated with a redistribution of the individual contributions to total lower-extremity work away from the ankle toward the knee and hip joint which was more distinctive in the recreational runner group than in the competitive runner group. This redistribution was accomplished by significant (P < 0.05) reductions of the external ground-reaction force lever arm and joint torque at the ankle and by the significant (P < 0.05) increase of the external ground-reaction force lever arm and joint torque at the knee and hip. Conclusions The redistribution of joint work from the ankle to more proximal joints might be a biomechanical mechanism that could partly explain the decreased running economy in a prolonged fatiguing run. This might be because muscle–tendon units crossing proximal joints are less equipped for energy storage and return compared with ankle plantar flexors and require greater muscle volume activation for a given force. To improve running performance, long-distance runners may benefit from an exercise-induced enhancement of ankle plantar flexor muscle–tendon unit capacities.


INTRODUCTION
Long-distance running is one of the most popular recreational activities in the world and is 49 often performed with competitive effort. High-performance runners differ from less successful 50 ones mainly in terms of the energy demand for a given submaximal running velocity, with 51 lower steady-state oxygen uptake indicating better running economy (1). Running economy is 52 a useful predictor of endurance running performance, which depends on a complex interplay 53 of factors such as the runner's training level, environment, anthropometric parameters, 54 physiology, and biomechanics (1). From a biomechanical perspective, running economy can 55 be related to spatio-temporal running characteristics (2), kinetics of the center of mass (CoM), 56 joint kinematics, and the tendons' capacity to store and return elastic energy (1,3,4). However, 57 no biomechanical parameter alone can explain the complexity of human running economy 58 (2,5). 59 Severe modifications of the running style, such as exaggerated knee flexion during the 60 stance phase (i.e., Groucho running), substantially reduce running economy by increasing 61 oxygen uptake (6). Reduction of running economy also occurs during sustained long-distance 62 runs performed until exhaustion (7,8). Fatigue, defined as exercise-induced reduction in the 63 ability to generate muscle force or power due to changes in the neural drive or exhaustion of 64 contractile function (9), can cause a decline in running velocity and changes in spatio-temporal 65 running characteristics and spring-mass behavior (10). However, whether these changes occur 66 when the running velocity is kept constant (as for instance during running on a treadmill) is 67 currently not clear (11)(12)(13)(14). Furthermore, despite one study indicating that knee flexion angle 68 at foot contact and mid-stance may be more flexed due to exhaustion on a treadmill (15), most 69 reports show relatively constant hip, knee, and ankle joint kinematics during prolonged 70 fatiguing treadmill runs (13,16). This appears to be independent of the performance level of 71 runners performing a 10-km treadmill run to volitional exhaustion at a velocity approximating 72 their 10-km race pace (17). 73 Only a few studies have examined the effects of exhaustion on running kinetics during 74 constant-velocity runs. In general, vertical ground-reaction force (GRF) and leg stiffness 75 decrease during exhausting running, whereas vertical stiffness tends to be rather constant 76 (11,14,18,19). However, considerable inter-individual differences seem to exist in the fatigue-77 induced changes in running kinetics (18,19). It is surprising that most reports investigating  Furthermore, well-trained distance runners with a good running economy show greater ankle 111 plantar flexor muscle strength and greater tendon-aponeurosis stiffness than runners with lower 112 running economy (4).

113
Although it is known that there are differences in individual joint contributions during 114 running, no studies have investigated if and how joint-specific work is altered over the course 115 of a prolonged fatiguing run (especially when performed at constant velocity) and whether 116 there are differences between recreational and competitive runners. The current study therefore 117 aimed to investigate the joint-specific contributions to the total lower extremity joint work 118 during a prolonged fatiguing run in recreational and competitive long-distance runners. The 119 primary hypothesis was that a long-distance run with near-maximal effort would change the 120 work contributions of the lower extremity joints, characterized by a reduction of work at the 121 ankle joint. A secondary hypothesis was that recreational runners would experience greater 122 running-induced reduction of ankle joint work than competitive long-distance runners. The    Kinematics and kinetics 158 The kinematics and kinetics were captured with 13 infrared cameras using a three-dimensional Step length, step frequency, and contact time were assessed for spatio-temporal 184 characterization of the running. Additionally, various kinematic and kinetic parameters were 185 determined during the stance phase of the right leg from the sagittal plane for further analysis 186 over the course of the run. To improve reliability, the data were averaged over 20 stance phases 187 at each of the 13 distance points (0 km, 0.2 km, 0.5 km, 1 km, 2 km, 3 km, 4 km, 5 km, 6 km,

218
Perceived exertion after the run was comparable between the two groups (RR: 16.9 ± 1.3; CR: showed a steady modest decrease over the course of the run (Fig. 2). In the RR group, there 241 was a slight increase in the positive joint work at the knee and hip joint, with statistically 242 significant (P < 0.05) increase at the knee joint at 2 km, 8 km, and 10 km. The further distance 243 points between 2 km and 8 km as well as the 9 km were slightly above the level of significance 244 (P > 0.05). In the CR group, the positive work showed a minor increase at the knee joint, but 245 did not change at the hip joint (Fig. 2). A significant (P < 0.05) intergroup difference but no 246 running distance main effect was seen for the total positive work of all three joints (Tab. 1).

247
Therefore, it is not surprising that the relative joint-specific contributions to the total positive 248 lower extremity joint work showed changes from the beginning of the run (hip 19%, knee 29%, 249 ankle 52%) to the end of the run (23%, 33%, 44%) in the RR group but not in the CR group 250 (beginning of the run: hip 22%, knee 28%, ankle 50% vs. end of the run: 22%, 31%, 47%). In the RR group, negative joint work was slightly increased at the hip, knee, and ankle over the 253 course of the run. In contrast, the negative knee joint work of the CR group was only slightly 254 increased and at the ankle slightly decreased (Fig. 2). There was no running distance main extending the GRF lever arms of the knee and hip joint (Fig. 4). These alterations in GRF lever 318 arms could explain the increases in knee and hip joint torques, as well as the decreases in the 319 ankle joint torque (Fig. 3). In this study, we found maximal torque magnitudes to be higher at the ankle joint 322 compared to the more proximal joints during running. In contrast, it has been reported that the and therefore internal Achilles tendon lever arm, this suggests that less force was acting on the 451 Achilles tendon, leading to a lower strain and hence decreasing energy storage in the tendon.

452
Accordingly, the increase in angular velocity in the CR group must originate from higher 453 muscle fascicle contraction velocity and not by a faster tendon recoil. We did not find a running 454 distance main effect for negative ankle joint work, which suggests that the runners were able the Achilles tendon, e.g. in long-distance running (34,48).

469
This study has several limitations. First, the individual season best times were self-reported, 470 and it is possible that the participants did not disclose their actual best times. Second, the 471 running economy was not directly quantified. Running economy has consistently been reported 472 to decrease during long-distance runs performed until exhaustion (7,8) and therefore it is very 473 likely that the participants of the present study also suffered from a reduced running economy.

474
In addition, we did not use spirometry because we speculated that wearing the spirometer 475 would affect running mechanics. Third, we did not determine the isometric or isokinetic force