Is Thoracic Spine Extension Work Necessary? – Part 1

About the Author: Eric Cressey

Today’s guest post comes from my friend and colleague, physical therapist Eric Schoenberg.  Eric is an integral part of our Elite Baseball Mentorships, and here, he kicks off a three-part series that I think you'll find very educational – even if it is a bit "geeky" along the way.  It'll test what you know and make you question some of the stuff you do with your training and work with clients. -EC

I’ll start by saying this is a classic case of giving an athlete what they need, not what we predict or assume they will need.  Let’s be clear up front: there are a lot of athletes that lack thoracic extension and rotation and certainly can benefit from T-spine mobility work.  This is especially true for your general population clients that live in flexion all day – although not every “average joe” is stuck in flexion; this is a wrong assumption!  However, for the purpose of this series, I am going to make a case for the following statement:

Not every individual – especially in baseball populations – needs thoracic extension and rotation mobility drills!  

In fact, to take it a step further, I would argue that in some cases, performing these types of exercises will actually make the athlete worse.

During our Elite Baseball Mentorship in August, I mentioned in passing during the postural examination portion of the breakout session that a lot of athletes present with flattened thoracic spines and some are actually relatively extended.  I followed this up by stating that some people don’t really need any T-Spine extension work.  This simple statement was met with a lot of surprise and follow up questions.

As a result, Eric and I thought it would be a good idea to explore this concept in greater detail.  As is often the case, a relatively benign statement takes on a life of its own and turns into a multi-part blog series.  As we develop these concepts, please feel free to share your thoughts and experiences in the comments section below.


The general assumption of the group was that everyone was kyphotic and T-spine extension was a default group of exercises that EVERY athlete needed.  This premise, however, does not take into account a host of key considerations, such as:

  • Defining T-Spine Extension. (part 1)
  • Anatomy and role of the Thoracic Spine, (part 1)
  • Scapulothoracic joint kinematics, (part 1)
  • Importance of a thorough static postural alignment and movement examination, (part 2)
  • Scapular position/prominence, (part 2)
  • Effect of the T-Spine on adjacent joints such as cervical spine, lumbar spine, pelvis, scapula, humerus, and ribcage, (part 2)
  • Concept of relative stiffness, (part 3)
  • Faulty motor control resulting in limited thoracic flexion (yes, lack of flexion!), (part 3)
  • Timing and amount of relative T-spine extension and rotation in the pitching delivery. (covered in part recently by Matt Blake and Eric C. here and here.)

174px-Gray_111_-_Vertebral_column-colouredIn an attempt to properly define thoracic spine extension with respect to the baseball player, we must first look at functional anatomy. The thoracic spine, with its rib attachments from T1-T10, is built for stability – most notably to protect the internal organs.  In addition, the T-spine has thinner intervertebral discs, a feature that adds to its relative inflexibility.  The sagittal alignment of the thoracic spine is kyphotic: 40 degrees in adults. (Neumann D.A. 2002).  With that said, we are not really talking about the T-spine being “extended”, but instead are talking about the relative amount of flexion that an athlete is in.  With that description, it’s important to appreciate that T-spine extension drills are working to put an athlete into an acceptable amount of flexion!  It is this flexion (or convexity) that provides a surface for the concave, ventral surface of the scapula to “float” on and create the scapulothoracic joint. (medial border of scapula and ribs 2-7).

For these reasons, we need to not just label an athlete as being “kyphotic” or having a “flat T-spine.”  Instead, we should attempt to define which segments of the T-spine have static or dynamic alignment issue and/or movement dysfunction and address them accordingly.

An important concept to look at is the T-spine is comprised of 12 vertebrae, so we will often find components of flexion, extension, and rotation within those 12 segments. Crosbie, et al. reports that the majority of T-spine extension occurs in the lower T-spine during overhead arm movement. This makes sense due to the thoracolumbar junction and the shape of the lower thoracic vertebrae.  The T10-T12 vertebrae are similar in size and function to the lumbar vertebrae.  T10 has a different rib articulation than T2-9 and T11/T12 have no facets on their transverse process for rib attachments (thus, increasing relative mobility).  So, more often than not, when we see anterior pelvic tilt, and lumbar extension compensation, we can assume that the lower T-spine is doing the same thing.  So what if, for the purposes of this article, we grouped the lumbar spine as T10-L5?


As a quick aside, it is recognized that the thoracic spine moves in three planes of motion.  More specifically, with respect to the frontal and transverse plane, with single arm elevation, the upper T-spine (T1-T5) has ipsilateral coupling of the lateral bending and rotation, whereas in the mid-lower T-spine T6-T12, we see contralateral coupling of lateral bending and rotation.  So in the frontal and transverse planes, the T-spine essentially works as two distinctly different subgroups.  This is important to consider when prescribing exercises to address a perceived movement impairment.  In addition, it is important to remember that the thoracic spine and hips (along with a stable lumbar spine) provide the rotational separation needed to excel in the game of baseball. However, for the purposes of this article, we will focus on the sagittal plane motion of flexion/extension.

As we continue to look at the sagittal plane, but shift our focus to T1, we see a smaller vertebrae (see picture above) that more resembles C7 (and other cervical vertebrae) than the caudal segments (T2-T9).  A case can be made that T1 (cervicothoracic junction) functions more like a cervical vertebrae than a thoracic vertebrae.

This leaves us with T2-T9: eight segments with an extreme mobility problem!  This is where we are really trying to mobilize “out of too much flexion” to get to an acceptable (neutral) position to allow for enough scapular (namely upward rotation and elevation) and glenohumeral motion to give us enough “space” to throw a baseball without causing an impingement.

So how can we be sure that we are following all these rules in three planes of motion when we are performing our T-spine mobility drills?  Are we in fact gaining motion in the correct segments within the context of their coupled motions?  Or, are we simply “mobilizing” what moves easiest and taking the “path of least resistance”?

Check back later in the week for part 2 of this series to read more about evaluation of static alignment and movement testing.  In addition, we will discuss how scapula position can fool us and (in part 3) how too many “abs” can be a bad thing.

In addition, if you’re interested in more information like this, we would love to see you at one of our Elite Baseball Mentorship, with the next one taking place in June. Click here to learn more.


Sign-up Today for our FREE Newsletter and receive a four-part video series on how to deadlift!