New insights into lumbar flexion tests based on inverse and direct kinematic musculoskeletal modeling.

Measurement of maximal lumbar flexion is considered to be a crucial element in the assessment of lumbar spine mechanics in situations as diverse as physiotherapy, orthopaedics, ergonomics, sport or aging. However, currently, there is no consensus on a reference test. This study aims to characterise five maximal lumbar flexion tests (four classical tests and a new, specifically-developed test designed to constrain pelvic retroversion) based on a three-dimensional, participant-specific musculoskeletal model. Twenty-six male and female participants performed the five tests. Movements were modelled in OpenSim to estimate change in length in lumbar, hamstring and gluteus muscles, together with lumbar flexion and pelvic tilt. These so-called "inverse" kinematic results were compared using a two-way ANOVA (sex×test). In a second step, lumbar muscle change in length was computed using a direct kinematic method. Lumbar flexion and lumbar muscle change in length were found to be greater when participants were in seated postures, with little pelvic retroversion. Female participants were observed to have less lumbar flexion than male participants (77±14° and 91±12°, respectively). Hip extensor muscles (hamstrings and gluteus) were fully stretched during each of the five tests. Our results highlight the specific roles of hamstrings, gluteus and lumbar muscles into reaching maximal lumbar flexion. Coupling inverse and direct kinematic methods proved to be a useful tool to enhance our knowledge of lumbar tests. Our findings help to characterise the role of the muscles involved in lumbar flexion, and we propose some recommendations for improving and standardising these tests.