Fitness Research update:
The interrelationship of the thorax and pelvis

A partnership between Australian Fitness Network, the University of the Sunshine Coast and the Australian Institute of Fitness, Fitness Research studies the populations, communities and environments related to the fitness industry, with the mission of improving the health of Australians through an improved body of fitness knowledge.

Research paper: The interrelationships of the thorax and pelvis under varying task constraints
Research team: Elias Delphinus, Dr Mark Sayers; University of the Sunshine Coast
Published: Ergonomics ; Vol. 56, No. 4, pp.659-666
Read more:

Introduction: Research has indicated that the majority of back injuries have involved axial trunk rotation (Manning, Mitchell, and Blanchfield 1984), a movement that is a fundamental component of normal daily activity. In addition, strong relationships have been reported between the incidence of back injury and axial trunk rotation movement patterns at end ranges of motion (Drake and Callaghan 2008). However, few studies have reported on the interplay between the thorax and pelvis in determining total trunk axial rotation.

The role of trunk rotation is best qualified when movement patterns are coupled, reflecting the three-dimensional (3D) complexity of the axial skeleton (Willems, Jull, and Ng 1996). The greatest diagnostic value for the detection of spinal instability is achieved by a neutral spine posture during passive axial rotation (Haughton et al. 2002). Small changes in flexion/extension postures were reported to have generated significant effects on axial twist moment/motion. Controlling or restricting the degrees of freedom (e.g. experimentation under artificial task constraints) may be limiting our understanding of the mechanisms that underlie functional human movement. This is particularly important when attempting to assess the interactions of segments within a functional unit.

When the main clinical interest is ROM, it is important to correctly define spinal segments that are relative to the task constraints and the study hypothesis. The anatomical frame of the thorax deforms considerably during large trunk motion, particularly in the transverse plane. Therefore, rotational errors are recorded commonly as a result of this deformation. Researchers have indicated that these errors are compounded if the thorax model definition includes acromion markers (Wu et al. 2005). The use of fewer segments may reduce errors as movements of adjacent segments are highly coordinated (Leardini et al. 2011).

The purpose of this study was to investigate the interrelationship between the thorax and pelvis during coupled movement patterns. The first aim of this study was to compare the axial rotational ROM achieved by the thorax and pelvis across four angles of trunk flexion from the hips. It was hypothesised that both segments would experience a reduction in axial rotational ROM as the degrees of forward trunk flexion increased. The second aim was to analyse the axial rotational ROM of the thorax when the pelvis was constrained by two contrasting conditions. It was hypothesised that the thorax would overcompensate for the lack of pelvic contribution to axial rotation due to the pelvic constraint conditions.

Methods: A randomised, repeated-measures experimental design was used to investigate the effects of four forward trunk inclinations (08, 158, 308 and 458) and two pelvic conditions (constrained and unconstrained) on maximum active rotation of the thorax and pelvis (Figure 1). An eight-camera motion analysis system (Qualisys, Medical AB, Gothenburg, Sweden) sampling at 250 Hz was used to record 3D movements of the thorax and pelvis. Retro reflective 14mm markers were attached to the skin overlying the 7th cervical and 10th thoracic spinous processes, the suprasternal notch and xiphoid process to model the thorax. The pelvis was modelled by attaching 14mm retro reflective markers bilaterally to the skin overlying the anterior superior iliac spine, posterior superior iliac spine and superior tip of the greater trochanters. Data was then modelled in 3D using standard biomechanical software (Visual3D, C-Motion Inc., Germantown, PA, USA) to construct a two-segment rigid body model of the thorax and pelvis (Wu et al. 2005).

Results: Figure 2 shows the total trunk axial rotational ROM together with the relative thorax and pelvic contributions to total rotation (%) at each angle of trunk inclination for the unconstrained condition. A mean 20° (SD = 2.58) reduction in axial trunk rotation for each 15° increase in trunk inclination was demonstrated. Pelvic axial rotation contributed more than 50 per cent of the total rotation up to 30° of trunk inclination. This pattern reversed at 45° of trunk inclination, where thorax axial rotation provided 54 per cent of the total rotation.

Figure 1. Representative photographs that display the passive marker locations used during data collection. A sample of the trunk inclinations and constraint conditions used during testing are included: (a) neutral (0°) constrained; (b) 15° unconstrained; (c) 30° constrained; (d) 30° unconstrained, and; (e) 45° constrained.

Summary: This study demonstrated that maximum trunk axial rotational ROM achieved under externally constrained conditions could provide rehabilitative protocols with unrealistic goals. For instance, if maximum trunk axial rotational ROM is used as a target for self-directed spinal rehabilitation, the likelihood for soft tissue micro-damage and exacerbation increases. The practical implications are simply that if trunk rotation is required, the pelvis should be allowed to rotate. For example, chairs must be allowed to swivel if a seated task requires rotation of the torso. Similarly, if torso rotation is required for a sporting activity the stance should be adjusted to allow the pelvis to rotate freely during ballistic rotational movements. In addition, flexion at the hips should be minimised if maximum rotation of the trunk is desired. The implications of these findings have direct relevance for individuals whose work tasks or sporting demands involve the coupled movement patterns of trunk flexion and rotation.

Conclusion: Thorax rotation was dependent on the level of constraint the pelvis experienced. It appeared that the self-organising nature of human movement, while in the endogenous condition, took priority over the potential axial rotational ROM available to the thorax. This suggests that the human musculoskeletal system is inherently protective.

Join Fitness Research’s Mark McKean to benefit from his scientific findings at WAFIC 2013:

• So you want to work with athletes? • B4E
• Fitness research updates • C1C
• Monitoring and management of recovery • C4B

For more information, session details and to register online, visit