ATTENUATION OF LEG IMPACTS DURING HIGH SPEED RUNNING

Abstract

Shock attenuation (SA) has been shown to increase as running speed increases (Shorten & Winslow, 1992) and as stride length (SL) increases (Derrick, Hamill, & Caldwell, 1998). Since SL has been reported to plateau at higher running speeds (Luhtanen & Komi, 1978), it was hypothesized that SA would change similarly. Subjects (n = 8; 25 ± 4.6yo; 80 ± 8.9kg; 1.79 ± 5.7m) ran at speeds equivalent to 50,60,70,80,90, and 100% of individual maximal speed with actual velocity calculated post-hoc. Order of speeds was counterbalanced, with ample rest between conditions to minimize fatigue. Light-weight accelerometers were secured to the forehead and distal aspect of the tibia. Data were collected (1kHz) for 20s with ten stance phase head and leg acceleration profiles analyzed per subject per condition. Head and leg power spectrum densities (PSD) were calculated for stance data and SA represented as the linear mean of 10log10(headPSD/legPSD) across the 10–20Hz frequency spectrum. Additionally, SL, leg and head impact peaks were recorded across ten impacts. Using repeated measures ANOVA with post-hoc curve fitting, SA, SL, leg and head impact were different across speeds (p < 0.05) and all increased linearly across speeds (p < 0.05). Higher order (2,4) polynomials also fit SL but this was most likely due to changes in treadmill speed which was fit with a 4th order polynomial (p < 0.05). Lack of an observed plateau of SL at high speeds suggests that maximum speed was not achieved by subjects. Since SL increased primarily in a linear fashion with speed (r = 0.92), similar relationships were observed between SL and SA (r = −0.71) and percent speed and SA (r = −0.61). However, for a two-fold increase in running speed, there was a 60% mean increase in SA, 53% mean increase in SL, and 30% increase in stride frequency. Despite nearly 5g mean increase in leg impact (50%: 6.1 ± 1.9g; 100%: 10.9 ± 2.8g), change in mean head impact was only 0.5g (50%: 1.4 ± 0.5g; 100%: 1.9 ± 0.3g). Changes in running speed were accomplished mainly by SL changes. Larger leg impacts across speeds were attenuated in such a way that magnitude of head peak impacts were maintained within an acceptable physiological range.