How the Scapula Generates Axial Rotation of the Humerus
Summary
Sufficient axial rotation range of motion is critical for performing such simple tasks as reaching into your back pocket (internal axial rotation), or combing your hair (external axial rotation). I previously described how axial rotation can be measured in a physically meaningful manner. In this project, I explore how scapular motion can generate axial rotation. This mechanism has not been previously described in biomechanics literature and provides a more comprehensive view of shoulder motion.
Code and Demos
| Scapula-generated Axial Rotation Results Repository | Python | |
| This repository contains code for reproducing the results of the publication associated with this project: Kinematic coupling of the glenohumeral and scapulothoracic joints generates humeral axial rotation. | ||
Current Status
- Complete
- Scapula-generated axial rotation analysis was published at the Journal of Biomechanics: Kinematic coupling of the glenohumeral and scapulothoracic joints generates humeral axial rotation.
Background
This project is a continuation of my prior project that explored how to measure axial rotation in a physically meaningful manner. The conclusion of that project was that axial rotation (rotation about the long axis of a bone) can only be measured correctly when considering the entire trajectory of a motion. This conclusion is not novel, and follows from the fact that 3D rotations belong to the mathematical group SO(3). However, almost all biomechanics literature incorrectly measures axial rotation due to the use of Euler angles.
My prior project and its associated paper provide more detail on why Euler angles are incapable of measuring rotations (although they are well-suited for measuring orientations). Euler angles assume a specific trajectory for the motion of the arm between any two points — one which is highly unlikely to result from natural human motion. The animation below visualizes this trajectory during a simple arm-raising motion (no scapulothoracic motion). See Appendix 1 of the prior project for an explanation of this motion path. In all subsequent animations, a red or black bar is utilized to represent the forearm. The black bar indicates the start of the motion, while the red bar indicates the end.
It's easy to see from the animation above that Euler angles assume a "strange" movement path between two endpoints. It's much more likely that the arm follows a simpler path between the endpoints, analogous to the animation below.
The start and end-point are identical between the two animations, only the trajectory of the arm changes. The Euler angle path (first animation) results in ~35° of internal axial rotation. The more direct path (second animation) — by design — results in 0° axial rotation. Recall that the main conclusion of my prior project was that the entire trajectory of a motion is needed to measure axial rotation correctly — as demonstrated by these animations.
In both animations the scapula does not move. What would happen if the scapula moved? Would that change axial rotation? Put another way: can movement of the scapula contribute to axial rotation? This post explores this question. Before continuing, I need to define some commonly utilized terminology that will facilitate this discussion.
The shoulder complex has a large range of motion because it is comprised of 4 sub-joints. In this post, I will focus on two of them: the glenohumeral joint and scapulothoracic joint (articulation is the correct term, but to ease the discussion I will refer to it as a joint). The scapulothoracic joint describes how the scapula glides on your ribcage. The glenohumeral joint describes how the humerus moves in relationship to the scapula. The combined movement of these two joints positions and orients your arm (humerus) with respect to the ribcage. Your arm range of motion would be severely limited if either your glenohumeral or scapulothoracic joint was inoperative. The motion of the arm with respect to the ribcage is described by the humerothoracic joint, and it is simply the combination of the scapulothoracic and glenohumeral joint movements — humerothoracic = scapulothoracic + glenohumeral.
It's easy to envision how the combined motion of these joints leads to humeral elevation (raising your arm). The scapula glides upon the ribcage to raise the humerus (see video below) — scapulothoracic motion leading to humerothoracic elevation. The humerus pivots on the scapula to rise higher — glenohumeral motion leading to humerothoracic elevation.
It's also easy to envision how the humerus can rotate around its long axis while the scapula remains still — glenohumeral motion leading to humerothoracic axial rotation.
However, it's more difficult to imagine how scapulothoracic motion can generate humerothoracic axial rotation. This is because we typically envision that the humerus and scapula are aligned as in the left side of the video below. For illustrative purposes, imagine that the humerus and scapula are aligned as in the right side of the video below. This alignment is non-physiological but it is useful to illustrate this idea. On the left side, every degree of scapular motion generates 1° of humerothoracic elevation. On the right side, every degree of scapular motion generates 1° of humerothoracic axial rotation.
Typically the alignment of the humerus and scapula is a mixture of the left and right configuration (e.g. think of having your raised arm resting midway between your front and your side). Therefore, during most arm movements the scapulothoracic joint (scapular motion) contributes to both humerothoracic elevation and axial rotation.
Let's revisit the prior animations. Recall that the Euler angle path measured ~35° of internal axial rotation, while the more direct path measured 0° of axial rotation. What if we allowed the scapula to move? In the animation below the the scapulothoracic joint contributes ~28° of axial rotation while the glenohumeral joint contributes none.
So which one is correct? Is there 35° of internal axial rotation from the glenohumeral joint, or 0° of axial rotation from the glenohumeral joint, or 28° of axial rotation from the scapulothoracic joint. None of them are correct, and all of them are correct. The answers above are correct for the respective paths of the humerus and scapula in each animation. However, all of the trajectories are fictitious — so none of them come from actual human subjects. I created these 3 trajectories to visually illustrate these points:
Euler angles are incapable of measuring axial rotation correctly. See the first animation for the "strange" movement path that Euler angles assume between two endpoints.
To measure axial rotation correctly we need to record the entire trajectory of the arm. See my prior project and its associated paper.
The scapulothoracic joint generates humerothoracic axial rotation during most arm movements.
In fact, we found that during arm elevation the scapulothoracic joint is the main contributor to axial rotation. The scapulothoracic joint also generates a significant amount of axial rotation during external rotation in abduction (with 15% of axial rotation originating from scapular motion and 85% from glenohumeral motion). You may read the accepted manuscript below.
Both axial rotation projects bring a fresh perspective towards understanding compensatory movement patterns. To understand the role of compensatory movement patterns in individuals with shoulder disease we must record the entire movement path and we must consider the role of the scapulothoracic joint in generating axial rotation. Finally, we must abandon Euler angles when calculating joint rotations — it is simply mathematically incorrect to utilize them for quantifying joint rotations.