The role of Transversus Abdominis (TrA) in lumbopelvic stability.

Common beliefs regarding the role TrA in segmental stability.

“Stability and control should be thought of as a dynamic process of controlling static position when appropriate in the functional context, but allowing the trunk to move with control in other situations” (Richardson, Hodges, & Hides, 2004, pp. 13-­14). Spinal stability is a concept that was derived in the early 1990’s and many physiotherapists have been taught the concept of stability through Panjabi’s model. This model “incorporates a passive subsystem, an active system and a neural control subsystem” (Richardson, et al., 2004, p. 15).

Along with the model of stability by Panjabi, physiotherapists’ are commonly taught that the muscular system is divided into global and local muscles, which act as two separate systems in controlling and providing movement, however, the evidence to support this distinct divide is low.

The local muscle system is often thought to comprise of deep muscles which have their origin and insertion directly onto the lumbar vertebra, including lumbar multifidus and tranversus abdominis. The global muscle system is therefore made up of the more superficial trunk muscles that cross multiple segments and attach indirectly to the vertebrae through fascial connections such as the thoracolumbar fascia (Richardson, et al., 2004).

Images above (Cleland, 2005, p. 140-143)

Retraining the active system has been the focus of much research over the past 20 years. “A critical feature of this research is that, in contrast to healthy control subjects, individuals with LBP do not demonstrate feedforward activity of the TrA during rapid limb movements” (G. T. Allison, Morris, & Lay, 2008, p. 228).

However, our understanding of the role of ‘stability muscles’ in the trunk is now changing.

“In 1996 Hodges and Richardson reported that patients with chronic low back pain (CLBP) suffered from inefficient stabilisation of the lumbar spine because they demonstrated a delay specifically in the activity of the transversus abdominis (TrA) during anticipatory postural adjustments associated with rapid arm movements. The authors reported that in healthy people TrA was the first trunk muscle activated in an anticipatory postural adjustment associated with unilateral arm movement and TrA activated independently of the direction of the perturbation to posture unlike the more superficial muscles (Hodges and Richardson, 1996; Hodges and Richardson, 1997b). As a consequence of this research, a theory developed regarding the specific importance of the TrA in stabilising the spine. TrA was purported to operate like a corset encircling, protecting and stiffening the core through bilateral co-­contraction (“corset hypothesis”) (Richardson et al., 1999). The “corset hypothesis” prior to movement was taught to patients, athletes and budding physiotherapists to enhance spinal stability both prophylactically and for those with CLBP” (S. L. Morris, Lay, & Allison, 2012, p. 249).

“Disagreement exists as to which muscles are most important to spinal stability and how these muscles most effectively provide stability while avoiding the negative consequences of excessive muscle activity (Brown, Howarth, & McGill, 2005; Brown & McGill, 2005; Cholewicki & VanVliet, 2002; Grenier & McGill, 2007; Hodges, 2001). While muscle co-­contraction clearly enhances spinal stability during lifting (van Dieën, Kingma, & van der Burg, 2003) it is unclear how spinal stability can be maximised during ballistic movements” (S. L. Morris, et al., 2013, p. 2).

Current research regarding TrA during rapid arm movement.

From more recent studies, spinal stability “can be interpreted to mean sufficient spinal stiffness to minimise unnecessary movement between spinal segments” (S. L. Morris, Lay, & Allison, 2013, p. 2).

In contrast to the above statement, TrA is now thought to be a part of a synergy of muscles contributing to axial rotation control associated with arm movements (Morris et al., 2012).

S. L. Morris, et al. (2012) conducted a study to determine the normal movement patterns of TrA associated with rapid arm movements. Their results rebutted the corset theory by proving EMG data that TrA does not engage bilaterally in preparation for movements, in fact it is the contralateral side which engages in anticipation for movement.

From these results, Morris and colleagues concluded that “it is not recommended that healthy people are trained to bilaterally co-­contract the TrA prior to movement as to do so would reinforce an unnatural strategy for movement” (2012, p. 252).

Feed forward activity of transversus abdominis (Morris et al., 2012, p. 253).    

Feed forward activity of transversus abdominis (Morris et al., 2012, p. 253).    

Morris, et al., (2013), are the first to demonstrate that TrA activity is closely related to global muscle activity in a whole body diagonal pattern. The results also demonstrated that there is a variability seen among individuals in motor control strategies.

Patients with CLBP have been reported to demonstrate less variable, more rigid bracing strategies that are not consistent with normal movement patterns. This study concluded, “TrA muscle activation reflects one element of the diagonal muscle synergy of muscle use associated with the efficient transfer of momentum from ground to hand. Bilateral co­contraction of TrA during asymmetrical arm movements is a rare motor pattern. The rationale for isolated co-­contraction TrA training in the prophylaxis of CLBP needs reconsideration” (S. L. Morris, et al., 2013, p. 9).

Muscle activation patterns with rapid right arm movement (Allison et al., 2008, p. 230).    

Muscle activation patterns with rapid right arm movement (Allison et al., 2008, p. 230).    

“Generally, there is an assumption that in the feedforward window (even prior to activation of other trunk muscles and anterior deltoid) the TrA is activated bilaterally and symmetrically and is not related to the direction of the perturbation—or, is not directionally specific” (G. T. Allison, et al., 2008, p. 234).

Allison and colleagues (2008) provide evidence to challenge the belief that the feed forward window acts bilaterally and that activity is dependent on direction of perturbation. A large proportion of health patients from this study demonstrate clear laterality in their TrA response to rapid arm movement.

Clinical implications

Some may believe that altered timing of the TrA may lead to poor spinal stability, however there is a lack of evidence to support this belief (G. T. Allison & Morris, 2008). Although it is plausible that bilateral activity of TrA will enhance inter-segment stability, this movement pattern has not been demonstrated during rapid arm movements in healthy patients.

Perhaps we need to reconsider how we verbally cue patients to engage their 'core muscles' prior to movement? This research provides a strong argument for observing patient movement patterns and assessing for a diagonal movement strategies for the control of axial rotation and lumbopelvic stability. 



Allison, G. T., & Morris, S. L. (2008). Transversus abdominis and core stability: has the pendulum swung? British journal of sports medicine, 42(11), 930‐931.

Allison, G. T., Morris, S. L., & Lay, B. (2008). Feedforward responses of transversus abdominis are directionally specific and act asymmetrically: implications for core stability theories. journal of orthopaedic & sports physical therapy, 38(5), 228-­‐237.

Morris, S. L., Lay, B., & Allison, G. T. (2012). Corset hypothesis rebutted— Transversus abdominis does not co-­‐contract in unison prior to rapid arm movements. Clinical Biomechanics, 27(3), 249-­254.

Morris, S. L., Lay, B., & Allison, G. T. (2013). Transversus abdominis is part of a global not local muscle synergy during arm movement. Human movement science, 32(5), 1176-­1185.

Richardson, C., Hodges, P., & Hides, J. (2004). Therapeutic exercise for lumbopelvic stabilization: Churchill Livingstone Edinburgh.