Exploring Biomechanical and Metabolic Determinants of Lifting Movement Strategy

Abstract

Background: Poorly designed manual materials handling (MMH) work, such as lifting, lowering, pushing, pulling, and carrying, can increase the risk of developing musculoskeletal disorders (MSD). To ensure MMH work is safely designed, digital human models (DHMs) can be deployed. A DHM enables a designer to simulate and understand the interactions between a worker and the work in a virtual environment, without the risk or expense of physical prototyping. However, our current understanding of how humans select movement strategies during MMH tasks limits our ability to accurately predict human postures in digitally simulated MMH environments. To address this limitation in our current understanding, more fundamental research is needed to uncover how biomechanics and energetics influence movement strategies during MMH work. Optimal Feedback Control (OFC) theory provides a comprehensive framework for explaining why individuals choose specific movement strategies. OFC proposes that people move in ways optimized for task-specific and situation-specific performance criteria, such as minimizing low back moments or metabolic cost. Currently, we lack a thorough understanding of the relevant performance criteria for MMH tasks like lifting, lowering, pushing, pulling, or carrying. By leveraging optimal feedback control theory as a framework to uncover biologically relevant performance criteria, in the future we can improve our ability to predict and simulate MMH movements. Purpose: The study was designed to determine the effect task parameters including load mass and lift frequency and time (pre- and post-exploration) have on the biomechanical exposures and metabolic costs of lifting. A research paradigm designed to study optimal control of gait was adapted to investigate the biomechanical and metabolic determinants of lifting movement strategy.

Methods: Using a repeated-measures experimental design, participants performed four 7-minute bouts of repetitive lifting in two different sessions, a high load low frequency (HLLF) session and a low load high frequency (LLHF) session. Within sessions, participants completed lifting bouts under 4-different technique conditions, where the first and fourth bouts allowed participants to self-select their technique (SS1, SS2) and the 2nd and 3rd bouts required lifters to adopt squat (SQ) or stoop (ST) techniques, respectively. High and low loads were defined as 15% and 5% of maximum voluntary lifting capacity in a semi-squat posture. High and low frequency were defined as 12 lifts per minute and 4 lifts per minute, respectively. Full body kinematics and VO2 consumption were collected during all trials. Using a whole-body top-down rigid link modeling approach, the peak sagittal L4/L5 moment was calculated. Two-way repeated measures ANOVAs tested for significant differences in biomechanical exposure and metabolic cost between the SS1 and SS2 lifting bouts.

Results: The group mean peak sagittal L4/L5 moment was 211 ± 7 Nm in the HLLF condition, and 149 ± 2 Nm in the LLHF condition. This task condition main effect was significant (F = 91.89; p < 0.001) where the HLLF condition resulted in significantly greater peak sagittal L4/L5 moments (64.6 ± 6.74 Nm; p<0.001). However, no main effect of time (F = 1.22 ; p = 0.28) or interaction effect was found (F = 0.46 ; p = 0.50). Significant main effects of task (F = 30.06; p < 0.001), and time (F =5.54; p < 0.05) were found for mean VO2 consumption, but no interaction effect (F = 2.81 ; p = 0.10) was found. Post-hoc analysis revealed that the LLHF condition resulted in significantly greater VO2 consumption (4.69 ± 0.86 mL.kg.min; p < 0.001), and that the SS2 technique had significantly greater VO2 consumption (0.95 ± 0.403 mL.kg.min; p < 0.05). An increase in VO2 consumption in SS2 was unexpected, so a secondary analysis was conducted to explore movement specific adaptations characterized using the Squat Stoop Index (SqStI). The SqStI analysis revealed that the mean for the last ten lifts across all lifting bouts was on average 13% greater in the LLHF condition relative to the HLLF condition (i.e. closer to a stoop lift). However, the change in SqStI between the SS1 and SS2 lifting techniques was 0.4 within the HLLF condition, and 1.3 within the LLHF condition.

Discussion: The magnitudes of the peak sagittal L4/L5 moments and relative VO2 consumption experienced by the participants were primarily affected by task parameters such as load mass and lifting frequency, which was expected by design. The biomechanical exposure and metabolic cost of participants did not significantly decrease following exploration of different lifting movement strategies. This was contrary to the hypothesis, where it was expected that VO2 consumption would decrease following exploration in the LLHF condition (i.e., learn to optimize for metabolic cost when the metabolic system is more challenged), and peak sagittal L4/L5 moments would decrease following exploration, but only in the HFLF condition (i.e., learn to optimize for biomechanical cost when the biomechanical system is more challenged). Instead, a significant increase in metabolic cost from SS1 to SS2 was observed across both task conditions. This could suggest a willingness to sustain a lifting movement strategy that increases metabolic cost over time in order to avoid increasing the biomechanical exposure experienced. Ultimately, participants were either unwilling or unable to significantly adapt their lifting movement strategy within the constraints of each task in order to reduce metabolic cost. A secondary analysis of lifting postures revealed that the initial and final preferred lifting movement strategy may have been modulated by the task parameters, where participants overall preferred a more stoop-like lift in the LLHF condition compared to the HLLF condition. In addition, it was seen that participants did not reach similar end ranges within the squat and stoop lifting bouts. Assuming that an SqStI value of 0% is a full squat lift, and an SqStI value of 100% is a full stoop lift, then participants were 31.6% away from a full squat but only 17.1% from a full stoop. This highlights a potential limiting factor of functional capacity, where participants may require additional relative strength, ankle range of motion, or aerobic fitness to explore a similar range of lifting movement strategies within the squat lift as compared to the stoop lift. Therefore, the results of this study did not support the hypothesis that an exploration of different lifting movement strategies would elicit movement strategy adaptation to optimize for certain task relevant performance criteria, however this may have been due to limiting factors such as the functional capacity of participants. However, the results of this study do demonstrate how varying task parameters can significantly modulate the biomechanics and energetics of occupational lifting, and thus the resulting preferred lifting technique. Although this study may not have uncovered how an individual’s optimal feedback control law may change during occupational lifting when exposed to different movement strategies, it does provide insight into how an individual’s preferred lifting movement strategy can be affected by the biomechanical and metabolic exposures experienced due to varying task parameters.