Glenohumeral Internal Rotation Deficit (GIRD) is a term used to describe to the apparent loss of shoulder internal rotation (IR) range of motion often seen in overhead throwing athletes. The loss of IR in the throwing shoulder is often accompanied by an increase in external rotation (ER) range of motion. There is a lot of speculation about the cause of these adaptations. A popular theory is that the loss of IR is caused by tightness in the posterior capsule of the shoulder and other soft tissue structures. Consequently, stretching the posterior shoulder is a commonly used treatment strategy. Given the uncertainty about the cause of GIRD and its relationship to injuries, it is worth considering that stretching the shoulder to increase IR may be the wrong approach.
Normal Range Of Motion For Overhead Athletes
Shoulder injuries are common in overhead athletes likely due to the large forces placed on the shoulder by the throwing motion. Overhead athletes require both flexibility and stability in their shoulders to meet the demands of their sport. Since these athletes typically have less IR range of motion (ROM) in their throwing shoulder compared to the opposite arm it is frequently thought that this adaptation is a contributing factor to injuries. A definite link between GIRD and injuries has however not been established.
When assessing IR ROM it is always important to also look at ER ROM to determine the total rotation ROM available at the shoulder. Below is an illustration showing the change in shoulder rotation ROM that is often found in throwing athletes. Both shoulders have the same amount of total shoulder motion (180 degrees), but the shoulder on the right has the typical overhead athlete presentation of increased ER and decreased IR.
The total ROM of the throwing arm needs to be compared to non-throwing shoulder to determine if there is a true loss of motion present. In most cases, despite the shift toward ER, the total amount of rotation ROM is within 5 degrees of the non-throwing arm. This means that an athlete with GIRD has not lost motion, only that their mid-point for rotation has shifted. GIRD should be considered a normal phenomenon that allows athletes the ER range of motion necessary to perform well in their sport.
Because GIRD represents of shift, rather than a true loss, of ROM it has been questioned if this adaptation is due to tissue “tightness” or if something else is at work. Recent studies have indicated that the ROM changes seen with GIRD are caused by changes in the bone rather than the soft tissue. Bony changes have been found on MRI and CT scans in the shoulders of throwing athletes in several studies. These adaptations are known as humeral and glenoid retroversion, and may be the result of the repeated twisting motion of the throwing action. Such changes can readily happen in young athletes whose growth plates are still open, but may also occur in adult athletes.
Stretching GIRD May Do More Harm Than Good
If GIRD involves bony changes, stretching aimed at increasing IR in a shoulder without a loss of total ROM not only doesn’t make sense, it could potentially put an athlete greater risk for injury. As physical therapist Mike Reinold suggests, attempting to increase shoulder IR through stretching in an athlete with GIRD, but symmetric total range of motion, may actually cause more harm because it would increase total shoulder motion beyond a “normal range”, and possibly create instability.
It is unlikely that stretching in the conventional sense can alter the shape or angle of bone anyway. I also think it is also unlikely that stretching or joint mobilization has any long-term effects on the joint capsule or other connective tissue surrounding the shoulder. Therapists in general overestimate the effectiveness of manual therapy and joint mobilization for increasing laxity of articular tissue (e.g. joint capsule) and connective tissue.
Does Stretching Really Work Anyway?
The Journal of Orthopaedic and Sports Physical Therapy recently published an article on ways to effectively stretch the tissue in the posterior shoulder. The authors of the study acknowledge that bony changes and postural adaptations may play a part in GIRD. The ultimate aim of article though is to provide effective ways to stretch the posterior tissue of the shoulder. The article recommends variations of the cross-body and sleeper stretches for increasing IR ROM.
When looking at stretching interventions we need to think about what is actually happening when we stretch tissue. The simple stress/strain, plastic deformation models that often get applied when discussing ROM cannot accurately account for the complexity of biologically active tissue, especially non-contractile tissue like joint capsules and ligaments. We only need to look at conditions like adhesive capsulitis and Dupuytren’s contracture, or the ligament laxity that occurs during pregnancy, to appreciate the complexity of connective tissue extensibility.
The posterior shoulder muscles control the distraction forces on the shoulder during the throwing motion. When these muscles are fatigued increased tensile forces are placed on the posterior capsule. The repetitive loading may lead to connective tissue proliferation which then leads to a stiffer capsule. At the cellular level, fibroblasts alter their gene and protein expression in response to tensile forces. It is suspected that repetitive loading can lead to increased Type I collagen synthesis and consequently a stiffer shoulder. If tensile forces from throwing are what cause the posterior capsule to become “tight”, I have a hard time understanding how tensile forces delivered by stretching would then also cause the capsule to loosen up.
In contrast to connective tissue, the muscles in the back of the shoulder are more likely to gain extensibility as a result of stretching. The muscles that limit IR include the posterior cuff and the posterior fibers of the deltoid muscle. Since these muscles function to restrain motion of the shoulder during the throwing action it’s worth asking if there is any benefit to lengthening them. It’s not clear if stretching these muscles has any long-term benefit. It’s also not certain that stretching can cause a permanent changes in IR ROM.
One study for instance found that both cross-body stretching, and stretching combined with joint mobilization increased IR ROM over a 4 week period. However, ROM returned to baseline 4 weeks after the intervention stopped. This could mean that continued stretching is needed to produce long-term effects. But it could also mean that stretching the shoulder has no long-term effects.
If the tissue of posterior shoulder becomes contracted or shortened from throwing, this may actually represent an adaptation to improve throwing performance. A “tight” posterior shoulder can also be a protective response that reinforces the glenohumeral joint to provide shoulder stability. Overhead athletes are found to position their scapula (shoulder blade) differently than non-athletes. This altered scapula positioning can allow throwers to generate motion from their scapula while preserving stability of the glenohumeral joint. So we can’t be certain that GIRD is a cause of injury, and it may turn out that it is actually the body’s way of protecting itself against injury. As mentioned earlier, attempting to increase IR at the glenohumeral joint without accounting for other factors could potentially cause more harm.
As GIRD is becoming widely accepted as a normal variant in throwing athletes, a new model of GIRD has been proposed that defines two types:
- Anatomical GIRD: Loss of shoulder IR less than 20 degrees with total rotation motion symmetrical between shoulders.
- Pathological GIRD: Loss of shoulder IR greater than 20 degrees with a corresponding loss of total rotation motion greater than 5 degrees
Essentially, pathological GIRD would be a shoulder with less IR ROM while also having less total rotation ROM in the throwing arm compared to the non-dominant side.
Nature Of Throwing Injuries
Even when considering this new definition of pathological GIRD we need to think about cause and effect. Do injuries occur because of this loss of motion, or does an athlete’s shoulder lose ROM as the result of injury? Attempting to stretch and regain total ROM may be treating the symptom of a problem rather than the actual cause of the injury. Just as bone spurs develop in areas under heavy stress but are not themselves a primary injury, a tight posterior shoulder may simply be a reaction to heavy loads on the arm. The actual cause of injury in throwing athletes is more likely a cumulative load on the shoulder that overwhelms the shoulder muscles’ capacity to resist those forces. A treatment approach aimed at managing load and strengthening the entire kinematic chain may prove more beneficial than stretching interventions in isolation.
I don’t mean to suggest that stretching has no utility in the treatment and prevention of shoulder injuries. One interesting study found that baseball pitchers lost passive IR ROM immediately after pitching, and this effect lasted upwards of 24 hours. From these findings we might reason that a consistent program IR stretching may indeed be important for preserving IR ROM and preventing soft tissue contracture.
For an athlete whose shoulder falls into the pathological GIRD category though, stretching alone may not be enough to regain lost motion. Because of the complex nature of connective tissue, a permanent increase in shoulder ROM may only be seen once the threat of instability is removed. A comprehensive program aimed at increasing overall strength and neuromuscular control may be needed to decrease the threat of injury, which then would encourage the tissue in the posterior shoulder to shift back to a “normal” level of extensibility.