GIRD Mike Reinold

What You Need to Know About GIRD: What It Is and What it Isn’t

GIRD Mike ReinoldGlenohumeral internal rotation deficit, or GIRD, continues to be one of the most polarizing topics in baseball sports medicine.  It has become so popular that even athletes and the general public know about GIRD, often exhibiting fear and anxiety with just the mention of GIRD.

How many times has a baseball player come back from the doctor with their head down saying, “I have GIRD,” as if the world has just ended?

I do not feel that everyone truly understands GIRD, how to assess GIRD, or how to treat GIRD.  There are a lot of theories and assumptions out there that may or may not be true.

Here is my take on GIRD, and it is not exactly how everyone would describe GIRD.


What is normal range of motion in an overhead athlete?

Before we can have any discussion on what is considered abnormal range of motion in the thrower’s shoulder, we should make sure we understand what is considered “normal” in overhead athletes.

Throwers have very unique adaptations from the demands of throwing. Numerous articles over the past 15 years have consistently shown that the dominant shoulder in overhead athletes exhibits an increase of external rotation and a subsequent decrease in internal rotation.  However, if you take the total rotation motion and combine ER and IR measurements the numbers are almost identical.

I remember when we first started discovering this phenomenon many years ago and it is uncanny how you find essentially the exact same total motion arc on both sides.

Here is an illustration of what this looks like.  In this figure.  You see the nondominant shoulder on the left and the dominant shoulder on the right.  You see the shift in the arc of total rotational motion, however if you break down the components of ER and IR, you see that both sides total 180 degrees.

GIRD glenohumeral internal rotation deficit

This adaptation has been shown in too many publications to list here, but I’ll add a few:

GIRD total rotational motion

This is a brief list but you can see that the total rotation motion on both the dominant and nondominant shoulders is almost identical in every study.  Statistical analysis revealed no significant differences in range of motion side to side.


Why the Adaptation in Shoulder Range of Motion?

Since the first discovery of this loss of internal rotation on the throwing arm, there have been several theories as to the specific reason for the adaptation.

The first theory centered around the fact that since there was a loss of internal rotation, there must be a subsequent tightness of the posterior capsule.  In actuality, this is really a long shot of being the isolated reason for the loss of IR, assuming a very specific cause for GIRD.

With so many potential factors contributing to the altered range of motion, it is surprising to me how popular this theory became.  If IR is less on the thrower shoulder, we now jump straight to recommending aggressive internal rotation and posterior capsular stretching.  I guess whatever theory comes out first gets the most traction and popularity!

The major flaw of the posterior capsule tightness theory is that it does not take into consideration the very specific increase in ER that is also seen in overhead athletes, let alone the fact that total rotational motion is still the same side-to-side.  If the posterior capsule was the cause of the loss of IR, would we then assume that the anterior capsule has loosened precisely the exact same amount to allow the exact same increase in ER as the posterior capsule does to restrict IR?

That sounds pretty unrealistic to me.

Shortly after the posterior capsule tightness theory was presented, many researchers took a more scientific look at what could be causing this very precise shift in the arc of motion in baseball pitchers and other overhead athletes. Several studies have now been published that have assessed boney changes that could be associated with GIRD.  Using both MRI and CT scans, it is now well documented that the humerus of the throwing arm is more retroverted than the nondominant arm.

What this means is that the actual bone of your upper arm torques and adapts.

Imagine twisting and wringing out a towel.  This is exactly what happens to the humerus during throwing while your growth plates are open.  The body, bones especially, do a great job of adapting to stress.  Essentially, the humerus bone of your upper arm twists at the growth plate and causes permanent adaptations to your bones.  Newer research is also now showing that the other end of the socket, the glenoid, also shows retroversion.

Based on these studies, the exact amount of retroversion observed appears to be approximately 10 degrees on average.  Now go back up and look at the table above.  Notice how the loss of IR is approximately 10 degrees and the gain of ER is approximately 10 degrees?  This boney adaptation makes the very specific shift in range of motion make more sense.  I have shown a simple and fairly effective way of measuring humeral retroversion in the clinic.  Try it in your throwers and you’ll see.

Pretty cool, right?  Still think the posterior capsule is the cause of loss of IR?

I was also a part of two studies that looked at glenohumeral translation in the baseball pitcher that both showed that posterior translation was twice that of anterior translation.  This was even present in baseball pitchers with as little as 10 degrees of IR.  They still had a large amount of posterior translation, not posterior capsular tightness.

Taking all of this into consideration, if there is one thing you take away from this article, it should be:

[quote]The thrower’s shoulder is supposed to have less IR on the dominant side.  This is normal.[/quote]


Determining What is Clinically Significant GIRD

A threshold to determine what can be considered a clinically significant loss of IR is vitally important to the implementation of programs designed to prevent and treat GIRD.  As previously discussed, a loss of IR itself can be considered a normal anatomical variation observed in overhead athletes.

Despite this finding, the term GIRD has continued to have a negative connotation, implying that any side-to-side loss of IR may be pathological.  This has resulted in a trend towards assuming many of the hypothesized theories of why loss of IR occurs are present in each person.

This unfortunately leads to a standardized prescription of stretches and exercises based on assumption and not a thorough assessment.

After reviewing the literature, it appears that most authors have been arbitrarily defining GIRD as a loss of IR greater than 15-20 degrees in comparison to the nonthrowing arm.  Some authors have even published studies showing that your chance of getting injured is increased if you have GIRD of more than 15-20 degrees.

Correlating GIRD to injury is too simplistic at best and again has too many flaws to consider this valid.

You can not accurately state why an increase in injury was observed.  Was it the loss of 17 degrees of IR?  Or perhaps the subsequent gain of 17 degrees of ER?  You can’t make a definitive conclusion either way.

[quote]Perhaps the increase of injuries in baseball pitchers is due to the gain in shoulder ER, not GIRD and the loss of IR?[/quote]

I am thinking this more and more everyday.

Another major flaw with defining GIRD using an arbitrary number is that the published amounts of range of motion have a very large standard deviation.  If you look through the published studies on GIRD, you’ll see that the standard deviation of measurements is large, ranging from 8 degrees to over 15 degrees.  That means the “normal” amount of internal rotation on a shoulder is approximately 50 degrees, but plus or minus 15 degrees.  Thus both 35 degrees and 65 degrees of internal rotation should be considered “normal.”

I can say that I have observed this first hand in professional baseball pitchers.  I have seen just as many players with 140 degrees of total rotational motion than I have with 200 degrees of total rotational motion.  Sure, this averages out to 170 degrees.  But not all baseball pitchers have 170 degrees of total rotational motion.

With such a large standard deviation and variability in measurements, assigning an arbitrary number to define GIRD is too simplistic.


A New Definition of GIRD

These findings have caused me to alter the way I define GIRD and stimulated me to propose a new definition of GIRD based on total rotational motion.

Previous definitions of GIRD based on arbitrary numbers have resulted in generalized treatment programs that are not specific or individualized enough to be utilized in clinical practice.

I propose that a loss of side-to-side IR is actually a normal anatomical variation in overhead athletes and should not be considered pathological GIRD unless there is a subsequent loss of total rotational motion in the dominant arm as well.  

This definition essentially takes the large variability in ROM that has been observed in athletes into consideration and allows for a more individualized approach to treating GIRD. So:

[quote]GIRD is a loss of internal rotation range of motion in the presence of a loss of total rotational motion.[/quote]

In this new definition of GIRD, a pathological condition of GIRD is defined as a loss of IR in the presence of loss of total rotational motion.

Lets looks at this as both an illustration and an equation.  In the figure below, you see the normal arc of motion in an overhead athlete, and to the right, an altered total rotational range of motion due to a loss of IR.

a new definition of GIRD

You can observe this yourself by assessing the specific range of motion measurements.  To calculate GIRD, use this equation:

GIRD = (Side-to-side difference in ER) + (Side-to-side difference in IR)

Here is an example of two baseball pitchers with a loss of IR:

  • Player 1 = (D ER 120 deg – ND ER 100 deg = +20 deg ER) + (D IR 60 deg – ND IR 80 deg = -20 deg IR) = 0 deg – Despite a loss of 20 degrees this is not pathological GIRD because total motion is the same bilaterally
  • Player 2 = (D ER 120 deg – ND ER 100 deg = +20 deg ER) + (D IR 35  deg – ND IR 80 deg = -45 deg IR) = -25 deg GIRD – This represents a pathological GIRD because both IR and total rotational motion are limited

I would suggest that the first player above should not be considered or even called GIRD, despite the fact that there is 20 degrees less IR on the throwing shoulder.  Because the total motion is the same, this is a normal adaptation in this athlete.

In fact, I would comfortably say in my experience that if you tried to reduce that 20 degrees loss of IR when total motion is symmetrical, you would essentially be increasing the total rotation motion and creating instability in an already vulnerable joint.

I believe this causes more injuries than it helps.

Because of the negative association that we have established with the word GIRD, I would propose that we stop calling everything GIRD and reserve GIRD for when it is truly pathological.  This helps clear up confusion.

I strongly feel that this new definition of GIRD takes the individual variability of range of motion as well as the total rotation motion into consideration and is a much more accurate calculation to base treatment recommendations.


Why is this important?

The goal of this article is to share my experience treating baseball pitchers.  I have rehabilitated 1000’s of injured baseball pitchers and managed 1000’s of healthy baseball pitchers.  This unique experience really opened my eyes to what is “normal” in baseball players.

When I first started working in Major League Baseball, I quickly found out that much of what I believe to be “facts” were not always accurate.  I’ll be the first to admit this.  If you only treat injured baseball pitchers, you start to assume that some normal adaptations may be pathological.

While my experience has been with baseball pitchers, this information can be extrapolated to all overhead athletes as these findings have all been established in other sports, such as tennis and handball.

There are far too many people who see a loss of IR and immediately label it GIRD.  Furthermore, there are far too many people who label any loss of IR as GIRD and blindly treat the posterior capsule.  I’m not saying that posterior capsular tightness does not exist, I am just saying it exists far less than we are diagnosing it and there are many other reasons that we need to consider before we start challenging the integrity of the stabilizing structures of the thrower’s shoulder.  I even dedicated an entire webinar to showing you 5 ways to gain IR without stretching the posterior capsule.

Blindly assuming GIRD is pathological, stretching the heck out of IR, and treating the posterior capsule is harmful.


Assess, Don’t Assume

Even with a pathological GIRD using the above equation, you can not assume you know why they have a loss of IR.

I have previously published a study showing that there is an immediate loss of IR after pitching.  We theorized that this was too acute to represent any changes in the capsule and most likely represented muscular stiffness from the eccentric trauma associated with pitching.  This was even more apparent when we also noticed that there was a loss of elbow extension, which is also subject to extreme eccentric forces during throwing.

You can’t assume they need to be stretched to gain more IR or that the posterior capsule is tight.  Maybe it is.  Maybe it isn’t.  Regardless:

Assess, don’t assume.

There is a specific way to assess the posterior capsule, which I will share in an upcoming post.  In fact I am going to write a series on how to more accurately assess GIRD, internal rotation, and the posterior capsule.

By changing the way we assess and define GIRD, we can start to more accurately understand what is happening to these overhead athletes and provide the best care possible.

  • Great post Mike! Blindly treating without truly assessing is never a good idea. Interesting that Borsa’s article showing increased posterior translation in throwers is often overlooked. I agree that the gain in ER is really more diagnostic than the loss of IR as to me it indicates some loss of stability. Even increased posterior cuff tightness may be more of a “reactive tightness” to such instability. As you alluded to, really comes down to stability vs mobility with stability often overlooked in throwers (and definitely females due to increasd laxity) even those who at first appear restricted in IR. Very informative and looking forward to your upcoming series.

    Seth Oberst DPT CSCS

  • Seth, thanks for the comment. I agree that the tightness can be from the act of throwing itself as well as protective of instability, great thought!

  • Ben

    Mike, in the case of the congenitally hyper-lax pitcher, who shows, for instance, 145 D ER/ 55 D IR, and 140 ND ER/ 80 ND IR (really unusual, i know) – My inclination is not to touch the supposed gird, as he seems to have stabilized a throwing shoulder that had far too much available IR. Is this the right line of thinking? I have an actual case like this, completely asymptomatic. He scores 9 on the beighton. If anything, the dominant shoulder is borderline unstable, right? Great article! Thanks.
    Ben Fairchild, MS, CSCS

    • Ben, in this case I would agree. I see people like this, the super laxity guys. Yes, focus should be on stability not mobility. 200 deg of total motion is a ton, i would look for instability on the left.

      This is definitely an outside the box example but these things happen. This is where feel comes in and I totally agree with your feel.

  • Mike, what are your thoughts on loss of internal rotation ROM in non-throwers who have had labrum problems such as SLAP. If we go by the traditional capsular pattern, external rotation should be most restricted but we are seeing that ER gets back to normal and IR does not. Have you seen this S/P SLAP repair? How does what we know about GIRD apply to non-throwers?

  • Hi Paul, GIRD in non-throwers is completely a non-issue. It’s not GIRD if they aren’t an overhead athlete with the anatomical adaptation.

    When you talk about surgery, especially a capsulolabral surgery, you are probably dealing with capsular tightness from the surgery. Totally different ball game, treat accordingly per usual techniques.

    You are probably starting to notice this but capsular patterns are not always straight forward, I rarely even consider them, just treat what you see!

  • Great and informative article Mike! I’ve seen my share of pitchers and overhead athletes and what you presented makes perfect sense. You mentioned something in the past about total rotational motion and it has stuck with me since then. I’ve been cautious not to label my patients with GIRD because the common definition of it suggests most if not all pitchers have it which has never made sense to me.

    • Thanks Chris, appreciate it and glad it helped!

  • Pingback: Lo que usted necesita saber sobre GIRD(Déficit Rotación Interna Glenohumeral): qué es y qué no es | ST Solutions()

  • Thanks Mike, I appreciate your experienced perspective. We often joke in our clinic about the potential of HIRD. Hip internal rotation deficit and it’s implications to the gait cycle, low back pain, etc. Any chance you have some insight on this “new” malady?

    • Same thing. Measure ER and IR bilaterally and add them up… spooky! This does alter the gait cycle but could be coming from pelvic alignment. Get the pelvis orientated and your IR magically appears!

  • Pingback: Strength and Conditioning Stuff You Should Read: 9/26/13 | Eric Cressey | High Performance Training, Personal Training()

  • Pingback: Therapy Reads 9/22 – 9/29/13 | ActivePTblog()

  • Tim

    It’s no secret that throwing decreases IR. Is it ok for a person with a lot of laxity to stretch to prevent this loss from occuring or will it simply not happen because the person is lax in the first place? I’ve had a SLAP repair 6 years ago and am a guy with a ton of laxity. Just want to know if I should stretch IR before throwing and after throwing to prevent any loss or do I not need it?

    • Tim, maybe “stretch” isnt the right word for you. You should regain and maintain your motion. You shouldn’t stretch to gain motion unless you need it. You can gentle stretch the muscles to warm up but don’t focus on trying to gain more motion if you already have full motion.

  • Pingback: What to Read: August-October 2013 | Orthopedic Manual Physical Therapy()

  • Pingback: GIRD or Total Range of Motion Deficit? « Board of Certification, Inc.()

  • Thanks Sir, again an eye opener article from you

  • Outstanding article, Mike. GIRD was addressed in undergrad (AT), but was not address in my musculoskeletal classes in graduate school (PT). Even when I learned of GIRD in undergrad, it’s clinical meaning was vague. I’ll be sure to keep all of this in mind this upcoming baseball season with my guys. Thanks, Mike.

  • Howdy, I just came back from the OS office to have my son’s shoulder looked at. The Dr. has gird as the diagnosis, has ordered a MRI and wants to start PT soon. He said that my son was very tight in his throwing shoulder (left) and that even if the MRI shows a small tear, treatment with begin with PT. I will admit, that I don’t understand all this science of what gird is, I am just a mom that wants to make sure I am doing the right thing for my son that loves to pitch for his H.S. team. Is PT the right step? Any advice is most welcomed.

    • Yes PT Debbie. You dont want surgery unless it is an absolute must. Send your PT these articles.

  • Mike , I am confused about tying to find the true total motion of one of the overhead athletes a recently assessed. DS ER 120 DS IR 25 and NDS ER 100 NDS IR 35. What is his true total motion? If I restore IR to the throwing shoulder it would further his asymmetry. Does the ND shoulder lose motion for any reason besides an injury? I would think he would need more total motion on the ND side as well. How do you know what the safe amount of motion to restore is? I’m baffled he is the first kid I have assessed with this type of asymmetry.
    Any help would be appreciated
    Thanks Sean

    • Sean, it is possible that his ER on the throwing side is too much. Hard to look at numbers and comment as your measurements are likely different than mine. I wouldnt work on IR, that is the whole concept of the article. Work on dynamic stability if his static stability is poor. Good luck

  • Hi mike,
    I am a competive swimmer and have been dignoised with GIRD? The doctors are worries about not being able to do the surgery until I am older because I am in the 7th grade I haw already have 5 moths of physical therapy. What is the average age for for GIRD surgery and what does it include? Thanks!

  • Ben Clarsen

    Hi Mike, great article! This is very much in line with our recent study of risk factors for shoulder injuries in handball players. We saw absolutely no relationship between GIRD (as traditionally defined) and injury. However, we did find a statistically significant association between reduced total rotational motion and injury.
    Here’s the paper:

    • Hi Ben, Yes I am aware. I totally agree. The operational definition is flawed. The problem is this leads to poor treatment decisions. Thanks for sharing your work!

  • Luke

    Hi Mike,

    I just bought your optimal shoulder DVDs with EC. What would you say about someone who has limited internal ROM in there non-dominant arm with both arms having the same ER? I’m talking about myself of course. I have had ultrasound say that I have supraspinatus tendonitis although I feel the pain more towards the back, near my infraspinatus? I recently started going to a physio in Toronto and have been getting shockwave and manual therapy with ER, PushupPlus exercises.


  • Hi Mike, I understand your reasoning with regards athletes who have been playing their sport for years, especially through the various growth spurts. But can this explain loss of internal rotation in older athletes who are relatively new to their sport? I see lots of older gentlemen who have come to surfing late in life, picked up a shoulder injury typically through overuse on top of poor form, which then often presents chronically with loss of IR. The basic hands on work to stretch/mobilise the joint capsule does increase their IR & their pain free range of movement. Is it more likely in this population that it is actually a capsular problem (as a result of an old, poorly rehab’d earlier injury)? Thanks, Jerry

Thanks for sharing! Follow me online for even more great content!

Send this to friend