1. Discussion With FD and SBST, the peak knee moment decreased after step 2 and then stayed constant for the remaining steps. Conversely, for BD the peak knee extensor moment increased after step 2. While the lowest peak knee moments were associated with SBSL, there were no apparent differences between those associated with SBST and FD. Peak knee extensor powers were significantly largest during FD and SBST and smallest in SBSL. In BD, peak powers were significantly smaller than those observed for FD and SBST but larger than SBSL. In terms of the interaction, peak knee power was largest during transitions at steps 1 and 5 and decreased in absolute magnitude thereafter, except during BD, where the trend was reversed. Discussion With FD and SBST, the peak knee moment decreased after step 2 and then stayed constant for the remaining steps. Conversely, for BD the peak knee extensor moment increased after step 2. While the lowest peak knee moments were associated with SBSL, there were no apparent differences between those associated with SBST and FD. Peak knee extensor powers were significantly largest during FD and SBST and smallest in SBSL. In BD, peak powers were significantly smaller than those observed for FD and SBST but larger than SBSL. In terms of the interaction, peak knee power was largest during transitions at steps 1 and 5 and decreased in absolute magnitude thereafter, except during BD, where the trend was reversed. Introduction Considerable research has been dedicated to understanding the demands imposed by stair ambulation (McFadyen & Winter, 1988; Reiner et al., 2001; Reid et al. 2007; Beaulieu et al., 2008). Peak knee moments and powers are larger in descent than ascent and level-walking, rendering descent a difficult mechanical task. In descent, lower limb muscles are required to generate force while lengthening, thereby dissipating gravitational energy and performing negative work. Stair descent imposes demand that require elderly and clinical populations to develop compensatory strategies to modulate the imposed loads, however, we know little about the mechanics of these strategies. The purpose of this research was to delineate differences between traditional forwards stair descent and two alternate patterns; forwards step-by-step and backwards descent. Only the peak knee extensor moments and powers during midstance will be considered here. During midstance the supporting leg’s knee extensor moment dissipates energy to control the rate of lowering during this critical period of single-support. Introduction Considerable research has been dedicated to understanding the demands imposed by stair ambulation (McFadyen & Winter, 1988; Reiner et al., 2001; Reid et al. 2007; Beaulieu et al., 2008). Peak knee moments and powers are larger in descent than ascent and level-walking, rendering descent a difficult mechanical task. In descent, lower limb muscles are required to generate force while lengthening, thereby dissipating gravitational energy and performing negative work. Stair descent imposes demand that require elderly and clinical populations to develop compensatory strategies to modulate the imposed loads, however, we know little about the mechanics of these strategies. The purpose of this research was to delineate differences between traditional forwards stair descent and two alternate patterns; forwards step-by-step and backwards descent. Only the peak knee extensor moments and powers during midstance will be considered here. During midstance the supporting leg’s knee extensor moment dissipates energy to control the rate of lowering during this critical period of single-support. Methods Healthy males ( n =9) and females ( n =9) participated in the study. All subjects were free of neurological and musculoskeletal disorders. Subjects performed five trials each of forwards (FD) and backwards descent (BD) and twenty trials of step-by-step (SBS) descent-- ten to analyze lead limb kinetics (STSL) and ten for trail limb kinetics (SBST). The order of the stair descent strategies were counterbalanced across subjects. Trials were initiated with the subject’s preferred limb and performed at self-selected pace. The staircase was instrumented with four force platforms embedded in steps 2 through 5 (step 2 & 5: Kistler 9286A; step 3 & 4: AMTI OR6-7-1000 & OR6- 7-2000). Step 1 was the top edge of the stairs. The stairs dimensions were 30 cm run and 20 cm rise. Force platform data were sampled at 200 Hz. Seven Vicon MX-13 cameras sampled displacement data at 200 Hz. Analog data were low-pass filtered with a 4 th - order Butterworth digital filter ( f c =10 Hz forces; 6 Hz for displacement data). Inverse dynamics analyses were performed according to the Newton-Euler formulation with Visual3D software. Data were ensemble averaged and subjected to 4×4 repeated- measures ANOVA with stair step and descent condition as factors for peak knee extensor moment and power and the work done. Methods Healthy males ( n =9) and females ( n =9) participated in the study. All subjects were free of neurological and musculoskeletal disorders. Subjects performed five trials each of forwards (FD) and backwards descent (BD) and twenty trials of step-by-step (SBS) descent-- ten to analyze lead limb kinetics (STSL) and ten for trail limb kinetics (SBST). The order of the stair descent strategies were counterbalanced across subjects. Trials were initiated with the subject’s preferred limb and performed at self-selected pace. The staircase was instrumented with four force platforms embedded in steps 2 through 5 (step 2 & 5: Kistler 9286A; step 3 & 4: AMTI OR6-7-1000 & OR6- 7-2000). Step 1 was the top edge of the stairs. The stairs dimensions were 30 cm run and 20 cm rise. Force platform data were sampled at 200 Hz. Seven Vicon MX-13 cameras sampled displacement data at 200 Hz. Analog data were low-pass filtered with a 4 th - order Butterworth digital filter ( f c =10 Hz forces; 6 Hz for displacement data). Inverse dynamics analyses were performed according to the Newton-Euler formulation with Visual3D software. Data were ensemble averaged and subjected to 4×4 repeated- measures ANOVA with stair step and descent condition as factors for peak knee extensor moment and power and the work done. IMPLICATIONS OF ALTERNATE STAIR DESCENT STRATEGIES ON KNEE BIOMECHANICS: BACKWARDS DESCENT IS LESS DEMANDING Tyler Cluff & D. Gordon E. Robertson, PhD, FCSB School of Human Kinetics, University of Ottawa, Ontario, Canada IMPLICATIONS OF ALTERNATE STAIR DESCENT STRATEGIES ON KNEE BIOMECHANICS: BACKWARDS DESCENT IS LESS DEMANDING Tyler Cluff & D. Gordon E. Robertson, PhD, FCSB School of Human Kinetics, University of Ottawa, Ontario, Canada Results Peak knee extensor moments showed significant step ( F (3,14) = 12.50, p <0.001) and descent condition main effects ( F (3,14) = 104.23, p <0.001) with a significant step×condition interaction ( F (9, 8) = 25.27, p <0.001). Figure 1 shows the interaction caused both main effects. Peak knee extensor power also exhibited significant by step ( F (3, 14) = 14.95, p <0.001) and condition main effects ( F (3, 14) = 172.00, p <0.001) and a step×condition interaction ( F (9, 8) = 17.88, p <0.001). Figure 1, however, shows the step main effect may be dismissed by the step×condition interaction. There was a significant condition main effect ( F (3, 14) = 171.8, p 0.05). Results Peak knee extensor moments showed significant step ( F (3,14) = 12.50, p <0.001) and descent condition main effects ( F (3,14) = 104.23, p <0.001) with a significant step×condition interaction ( F (9, 8) = 25.27, p <0.001). Figure 1 shows the interaction caused both main effects. Peak knee extensor power also exhibited significant by step ( F (3, 14) = 14.95, p <0.001) and condition main effects ( F (3, 14) = 172.00, p <0.001) and a step×condition interaction ( F (9, 8) = 17.88, p <0.001). Figure 1, however, shows the step main effect may be dismissed by the step×condition interaction. There was a significant condition main effect ( F (3, 14) = 171.8, p 0.05). Biomechanics Laboratory References Beaulieu FG, Pelland L & Robertson DGE (2008) Gait Posture 27:564-571. McFadyen BJ & Winter DA (1988) J Biomech 21:733-744. Reid SM, Lynn SK, Musselman RP & Costigan PA (2007) Med Sci Sports Exerc 35:2005-2011. Reiner R, Rabaffetti M & Frigo C (2002) Gait Posture 15:32-44. References Beaulieu FG, Pelland L & Robertson DGE (2008) Gait Posture 27:564-571. McFadyen BJ & Winter DA (1988) J Biomech 21:733-744. Reid SM, Lynn SK, Musselman RP & Costigan PA (2007) Med Sci Sports Exerc 35:2005-2011. Reiner R, Rabaffetti M & Frigo C (2002) Gait Posture 15:32-44. Figure 1. Body-mass normalized peak knee extensor moments and powers and the work done at each stair during forwards, backwards and step-by-step stair descent (lead and trail limbs). Error bars show 95% confidence intervals. Summary and Conclusions Peak knee extensor moments and powers were relatively constant after one complete step cycle except for the leading leg of SBS descent, which had the lowest work, moments and powers. Work done, peak knee moments and powers were all significantly larger in FD and for the trail leg during SBS descent than for BD and SBSL. Therefore, both the BD and SBS compensatory strategies reduce the demands at the knee during stair descent. Backwards stair descent can thus be recommended as an alternative to forwards descent for individuals with bilateral weakness of the knee extensor musculature. Step-by-step descent only reduced the knee work, moments and powers for the leading leg and therefore is a feasible strategy for individuals with unilateral knee extensor weakness. Summary and Conclusions Peak knee extensor moments and powers were relatively constant after one complete step cycle except for the leading leg of SBS descent, which had the lowest work, moments and powers. Work done, peak knee moments and powers were all significantly larger in FD and for the trail leg during SBS descent than for BD and SBSL. Therefore, both the BD and SBS compensatory strategies reduce the demands at the knee during stair descent. Backwards stair descent can thus be recommended as an alternative to forwards descent for individuals with bilateral weakness of the knee extensor musculature. Step-by-step descent only reduced the knee work, moments and powers for the leading leg and therefore is a feasible strategy for individuals with unilateral knee extensor weakness. Acknowledgements T.C. was supported by an NSERC CGS-M scholarship. Thanks to F. Beaulieu for technical support. Acknowledgements T.C. was supported by an NSERC CGS-M scholarship. Thanks to F. Beaulieu for technical support. -5.0 -4.0 -3.0 -2.0 0.0 1.0 2345 Step Peak knee power (W/kg) -1.5 -1.3 -1.1 -0.9 -0.7 -0.5 -0.3 -0.1 0.1 0.3 0.5 2345 Step Knee work (J/kg) -0.5 0.0 0.5 1.0 1.5 2.0 2345 Step Peak knee moment (N.m/kg) Forwards Backwards SBS Lead SBS Trail