Lumbar spine marker placement errors and soft tissue artifact during dynamic flexion/extension and lateral bending in individuals with chronic low back pain

Quantitative in vivo biomechanical assessments are typically performed with optoelectronic motion capture (MoCap) using retroreflective markers attached to skin. This technique inherently contains measurement errors from both marker placement on palpated bony landmarks and skin motion relative to the underlying bone (i.e., soft tissue artifact (STA)). Research on lumbar spine STA is scarce and limited to young, healthy participants in static positions. This study aimed to evaluate static placement errors, lumbar spine STA from MoCap marker clusters (MMC), and linear relationships between STA and patient characteristics. Thirty-nine participants with cLBP performed three trials each of flexion/extension and lateral bending while imaged simultaneously by MoCap (120 Hz) and dynamic biplane radiography DBR (20 Hz). MMCs were placed 29.5 +/- 18.0 mm and 27.1 +/- 13.4 mm superior to the most prominent aspect of the L1 and L5 spinous process, respectively. L1 relative to L5 STA was larger during flexion/extension (8.6 +/- 5.7 degrees) than lateral bending (4.5 +/- 2.1 degrees) (p < 0.001). After correcting for marker placement errors, components of the L1 and L5 STA averaged as much as 16.3 mm and 11.4 degrees during flexion/extension, but only 4.0 mm and 4.8 degrees or less during lateral bending. On average, STA for individual L1 and L5 vertebrae increased as participants moved away from the upright neutral position. STA was participant-dependent, however, age and BMI did not model STA well. Given the inaccuracy in marker placement and wide range of patterns of STA, caution is urged when making clinical decisions or when using computational models to estimate spine tissue loading based upon lumbar spine kinematics obtained from skin-mounted markers.