PT on the Net Research

Four Point TVA Exercise


I have a question about breathing while your TVA is engaged. While breathing properly, the belly should rise and fall. While engaging the TVA, the belly doesn't move while breathing, correct? So while performing the TVA drawing in exercise on all 4s, the instructions say "draw in the belly, and continue natural breathing." In doing so, do we then rely on accessory respiratory muscles to breath as opposed to the diaphragm? I was hoping you could explain how breathing changes during TVA activation (the belly no longer rises and falls?) and could provide tips on how I could instruct my clients to breath correctly while engaging the TVA.


This is a great question and one which deserves a comprehensive answer as it is a source of confusion and misunderstanding in many sectors of the fitness and rehabilitation industry.

The TVA is primarily a muscle of stabilization and secondarily a muscle of respiration (Lewit 1999). The 4 point TVA exercise is a very “non functional” exercise and, as such, is a useful way to isolate the TVA (for many reasons), but it is always an exercise that we should be looking to progress the client beyond as rapidly as possible.

If the TVA is being engaged isometrically, as in the 4 point TVA, it is consciously being used to stabilize the spine, and its respiratory function is switched off. In this instance, the client would have to use the accessory muscles of respiration, as you correctly deduce. However, using the accessory muscles is a dysfunctional breathing pattern that leads to all kinds of detrimental physiological ramifications (Chaitow 2004) and therefore is not desirable in the longer term.

If a client finds that they can’t breathe at all while performing a 4 point TVA or similar exercise, it is an indication that they are recruiting their obliques as well, which serves to pin down the rib cage, making even accessory respiration impossible.

Interestingly, Urquhart and Hodges (2005) found that the lower abdominal wall fires ahead of the middle and upper abdominal wall which, by implication, tells us that the lower abdominal wall is primarily tonic in function. We can deduce this because tonic motor neurons have a lower threshold to stimulus, which is why muscles with a tonic (or postural) dominance switch on ahead of those with a phasic dominance; the “feed forward mechanism” of stabilization before mobilization (Richardson, Jull, Hodges & Hides 1999).

What we know of pelvic stabilization, such as the force and form closure mechanisms (Vleeming 1989, Lee 2004) and, in particular, the “nutcracker phenomenon” (where the ilia are like the handle of the nutcracker, the sacrum like the nut, the posterior sacroiliac ligaments like the hinge and the infra-ubilical, inter-ASIS portion of TVA like the hands), which highlights the importance of good tonic function in the lower abdominal wall to provide a stable base for the head arms and trunk to function from.

It is for this reason that a current train of thought is to allow the upper abdominal wall to still move during the respiratory cycle (thereby still allowing diaphragmatic excursion), while maintaining an isometric contraction of the lower abdominal wall (McNeill 2009), a kind of upper/lower abdominal dissociation. This is, it would appear, a useful interpretation of how the abdominal wall is designed to work.

The old Pilates concept of “lateral breathing” while keeping a flat tummy is now considered by most leading thinkers debunked and defunct (Lee 2004, McNeil 2009), creating far more problems than it does solve them (and in particular, inverted breathing patterns, subsequent ANS imbalance and visceral dysfunction).

To see how an abdominal wall is supposed to work, watch a young child breathe. At this stage of the development, the child usually has no history of back pain, no significant visceral issues (though this is becoming more commonplace) and no social inhibitions with regard to his appearance. The idea of a flat stomach being functional is, in fact, an adult projection of what function should be. Certainly the stomach should engage and go flat under circumstances where stability is required (again, watch a toddler or young child move or jump... his abdomen draws straight in), but if it remains flat and static after that stability challenge has gone, it is a dysfunctional abdominal wall being held under conscious control and/or faulty motor programming.

However, abdominal wall function and respiratory integration was NEVER SUPPOSED to be under conscious control. It was always supposed to “just work.” The problem is that the incidence of back pain and of visceral dysfunction (both of which reflexively mess with firing patterns) is so high that you will be hard pressed to find anyone beyond the age of 10 whose abdominal wall “just works.”

What is useful to remember is that the goal of abdominal conditioning is to return the client to this exact point, where the abdominal wall “just works.” This means it is has to be a whole health approach and also means that, ultimately, clients shouldn’t need to think about their abdominal wall or their breathing. They should be an “unconscious competent” (Brandon 2006). Indeed, much research suggests that the TVA should ONLY function at one to three percent MVC (McGill 2001). Try doing that consciously... it’s not possible! You simply cannot consciously only contract to that minimal level.

Further, an understanding of length/tension relationships, in conjunction with recognition of the thoracolumbar fascia’s contribution to spinal stability and the difference between isometric, concentric and eccentric contraction, allows you as the specialist to coach your clients that, so long as their abdominal wall moves out, to the same extent that their diaphragm moves down, the length/tension relationship in the thoracolumbar fascia remains at a constant. This means that lumbo-pelvic stability also remains at a constant AND the viscera get massaged, AND the pressure differentials created draw fluids back from the periphery through the venous and lymphatic systems, all as part of the same bargain! And on top of this, if the abdominal wall does move out as the diaphragm moves down (contracts concentrically), this DOES NOT mean that the transversus abdominis has “switched off,” but that it can be (hopefully “is”) working eccentrically. In this way, both the length/tension relationship and the intra-abdominal pressure can be maintained at an approximate constant. A kind of “intra-abdominal/thoracolumbar fascial/lumbopelvic stability-homeostasis.”

This arrangement helps to explain how the spine and pelvis retain stability while we walk or run, for example (where ongoing breathing is required). In short, you’re right about using accessory muscle with complete engagement of TVA in the 4 point TVA exercise. However, you may want to focus on maintaining an isometric contraction of the lower abdominal wall and allowing upper abdominal wall excursion (the upper abdominal wall is more phasic anyway), and ultimately, all this should occur without thinking. An isometrically held flat tummy is only healthy if it is under load.


  1. Chaitow L, (2004) Breathing pattern disorders, motor control and low back pain. JOM 7(1)pp33-40
  2. Lee, D (2004) The Pelvic Girdle (3rd Edition). Pub: Churchill Livingstone
  3. Lewit, K (1999) Manipulative Therapy in Rehabilitation of the Locomotor System (3rd Edition). Butterworth-Heinemann. pp27-28
  4. McGill, S (2001) Achieving spinal stability: blending engineering and clinical approaches. In “Proceedings of 4th Interdisciplinary World Congress on Low Back & Pelvic Pain”, pp203-210
  5. McNeil W, (2009) From Journal Page to Treatment Space. Journal of Bodywork & Movement Therapies. Volume 13, Issue 1, January 2009, Pages 93-97
  6. Richardson C, Jull G, Hodges P, Hides J (1999) Therapeutic exercise for segmental spinal stabilization. Churchill Livingstone
  7. Urquhart D, Hodges P, Storya I, (2005) Postural activity of the abdominal muscles varies between regions of these muscles and between body positions. Gait & Posture Vol 22, Issue 4, Dec 2005 pp295-301