The Shoulder

01What the surveillance shows

The Numbers on a Shoulder Under Overhead Load

Eight findings that move the overhead shoulder from a failing joint to a load path losing its distribution.

02Mobility bought with bone

The Shoulder Trades Bony Stability for the Largest Range in the Body

The shoulder gives up bone to buy range, and that trade defines every injury an athlete will ever have here. The glenoid fossa is a shallow dish rather than a cup. Measured across 25 anatomic specimen shoulders, the glenoid articular surface and the labrum together form a socket roughly 9 millimeters deep from top to bottom and 5 millimeters deep front to back Howell 1989.

The labrum supplies about half of that depth. Detach it at the front, as a Bankart lesion does, and the front-to-back depth of the socket falls from around 5.0 millimeters to 2.4. That is the whole margin an overhead athlete is working with, and it is made of tissue rather than bone.

Anatomists compare the result to a golf ball resting on a tee, and the comparison earns its place because it names the problem exactly. There is no bony wall to run into. Whatever holds the ball on the tee has to be holding it actively, every instant the arm is moving.

So the security of this joint is muscular. Two mechanisms carry it. Concavity compression is the cuff pressing the humeral head into the concave glenoid, and it gets stronger as the compressive load rises and as the concavity deepens. Scapulohumeral balance is the scapula being positioned so the joint reaction force stays inside the glenoid arc Lippitt 1993.

Both mechanisms depend on an intact labrum, and both matter most in the middle of the range, where the capsule and the glenohumeral ligaments are slack and holding nothing. The end ranges have ligaments. The working range has muscle and timing.

Why the cuff decides whether a joint is loose or free

This is why the shoulder cannot be understood as a passive hinge. Stability here is a verb, produced moment to moment by muscles responding to load, position, and intent. When that active control is precise, the athlete feels nothing and simply performs.

When it lags by milliseconds, the head drifts, soft tissue takes load it was never meant to carry, and the cascade toward impingement, labral injury, or frank instability begins. The cuff is the difference between a joint that is loose and a joint that is free.

The trade also fixes what a shoulder problem is. A joint whose security is distributed across muscle, labrum, capsule and scapular position does not fail at a point. It loses its distribution. That is the claim this page builds, and the scapula is where the loss shows up first.

03Four muscles, one force couple

The Rotator Cuff Centers the Head While the Deltoid Moves the Arm

The rotator cuff is four muscles working as a single dynamic stabilizer, and its job is to keep the humeral head centered while the big movers generate power. Supraspinatus initiates abduction and compresses the head down into the glenoid. Infraspinatus and teres minor produce external rotation and resist the head translating forward.

Subscapularis is the lone internal rotator of the four and the front wall against the head sliding anteriorly. Together they form a force couple with the deltoid, which is what lets the deltoid pull the arm up without the head riding up under the acromion with it.

What happens when the cuff gets tired

Fatigue the cuff and the head moves. Investigators radiographed the joint at 0, 45, 90 and 135 degrees of elevation in the scapular plane, before and after a task designed to exhaust the whole cuff. The fatigued state showed superior migration of the humeral head Chopp 2010.

The pattern is worth reading closely. Before fatigue, the head translated upward until about 90 degrees of elevation and then settled back toward center. After fatigue, translation kept increasing at the higher elevations instead. The centering function did not weaken evenly. It stopped switching over.

That is the subacromial space narrowing from the inside, produced by a muscle that ran out of capacity rather than by a bone that changed shape. The overhead athlete meets that state late in a set, late in an inning, and late in a season.

What one throw asks of the joint

The loads are large and they arrive twice. Motion analysis of 26 highly skilled adult pitchers found 67 newton meters of shoulder internal rotation torque and 380 newtons of anterior force shortly before the arm reached maximum external rotation Fleisig 1995.

Then, shortly after ball release, the same shoulder produced 1,090 newtons of compressive force, 400 newtons of posterior force and 97 newton meters of horizontal abduction torque. The authors name the cuff's own contribution to those loads as a route to tensile failure of the cuff itself.

Read the timing of those two instants. The first sits at the end of late cocking, where the shoulder is driven into extreme external rotation. The second sits just after the ball leaves the hand, where the posterior cuff has to fire eccentrically to decelerate a rotating arm. Braking is the harder job, and it is the one that fails.

They also name the labral consequence. Compression, joint laxity and that 380 newtons of anterior force during arm cocking can tear the anterior glenoid labrum, and rapid internal rotation on top of them grinds it. Biceps tendon tension, needed for both elbow flexion torque and shoulder compression, can tear the anterosuperior labrum.

So tendinosis, internal impingement and superior labral pathology sit downstream of a cuff that lost its timing rather than a cuff that simply got weak. Restoring the timing is the work. How a tendon behaves once it is painful belongs to Tendon Pain, and the energy that arrives at this shoulder from the legs and trunk belongs to The Kinetic Chain.

04Scapular dyskinesis

Six in Ten Overhead Athletes Carry Altered Scapular Control

Scapular dyskinesis is the ordinary state of the overhead shoulder rather than a rare defect. A systematic review of 12 studies covering 1,401 athletes, of whom 1,257 played overhead sports, put its prevalence at 61 percent in overhead athletes against 33 percent in nonoverhead athletes Burn 2016. The gap was significant below a P value of .0001.

The term names abnormal dynamic control of the shoulder blade on the rib cage, and it is identified by watching the scapula move rather than by imaging it. The reviewers are explicit that it appears in people with no symptoms at all, which is why a prevalence figure that high is not a catalogue of injuries waiting to happen.

It does carry risk, and the risk runs forward in time. Five prospective studies pooled 419 athletes who were free of shoulder pain at baseline. Those with scapular dyskinesis went on to a shoulder pain event 43 percent more often over the following 9 to 24 months, at a risk ratio of 1.43 Hickey 2018.

Pain arrived in 35 percent of the athletes with dyskinesis and 25 percent of those without. The finding matters because it is prospective. The scapula was moving differently before the shoulder hurt, which rules out the reading that dyskinesis is only a consequence of pain.

What the scapula does in a shoulder that already hurts

Once a shoulder is symptomatic the scapular pattern is specific and measurable. Fifty-two construction workers with routine overhead exposure were tracked with electromagnetic sensors and surface electromyography during humeral elevation in the scapular plane Ludewig 2000. They lifted nothing, then 2.3 kilograms, then 4.6.

The group with impingement symptoms showed decreased scapular upward rotation early in elevation, increased anterior tipping late in elevation, and increased medial rotation of the scapula under load. Upper and lower trapezius activity increased at the same time.

Read that pairing carefully, because it is the argument of this page in one experiment. The scapula was not idle and the muscle was not quiet. Effort went up while position got worse. A shoulder in trouble is not a shoulder doing less work. It is a shoulder distributing work badly.

Fatigue one muscle and the whole plate moves

The clearest demonstration puts the fatigue in one place and finds the change in another. Twenty-five overhead athletes with healthy dominant shoulders performed a low-resistance, high-repetition prone external rotation protocol, arm abducted to 90 degrees, moving from 0 to 75 degrees Joshi 2011.

The fatigue was applied to the glenohumeral external rotators. Afterward, lower trapezius activity fell in both the ascending and descending phases of a diagonal task, infraspinatus activity rose in the descending phase, and scapular upward rotation increased.

Lower trapezius does not cross the glenohumeral joint. It runs from the thoracic spine to the scapular spine, and it went quiet because a muscle downstream of it got tired. The tension changed at the humerus and the answer showed up at the shoulder blade.

The Unified Model of Tone states the principle in one line: a local change in tension does not stay local. The scapula is where the shoulder's load path publishes its accounts, and that is why it reads the state of the whole limb rather than the state of one muscle.

05The rotation budget

Internal Rotation Deficit Did Not Predict Shoulder Injury, and Too Little External Rotation Did

Glenohumeral internal rotation deficit is the most measured number in the throwing shoulder, and the largest prospective test of it came back negative. Investigators performed 505 examinations on 296 professional pitchers over 8 consecutive seasons, measuring passive rotation with a bubble goniometer at spring training Wilk 2015.

Seventy-five shoulder injuries and 20 surgeries followed, across 51 pitchers and 5,570 days on the disabled list. Internal rotation deficit, total rotation deficit and flexion deficit were not significantly related to shoulder injury or to surgery.

One motion measure was. Take a pitcher whose throwing shoulder has less than 5 degrees more external rotation than the other side. That pitcher was 2.2 times more likely to reach the disabled list with a shoulder injury, and 4.0 times more likely to need surgery on it.

The deficit itself is defined as a loss of 20 degrees or more of internal rotation in the throwing shoulder against the nondominant one. An earlier case series of the same organization found pitchers with that deficit nearly twice as likely to be injured without reaching significance, and Orthopedic Screening, Upper Extremity carries the screening figures in full.

The budget moves on both sides

Part of why the deficit measures poorly is that the comparison shoulder is not standing still. Forty-eight healthy collegiate baseball and softball athletes were measured three times across one season, before the fall, before the spring and after it Dwelly 2009.

External rotation increased in the throwing shoulder over the season while internal rotation did not change. The total arc grew by roughly 9.7 to 11.0 degrees. External rotation increased in the nonthrowing shoulder too, and the total arc grew there as well.

No change in the deficit itself appeared over the season, and the two accepted ways of calculating it, the raw internal rotation difference and the percentage of total arc, flagged different athletes. A number that changes identity depending on which formula produced it is a number to read alongside others rather than alone.

Where posterior tightness actually shows up

Tightness at the back of the glenohumeral joint does not stay at the glenohumeral joint. Ten shoulders from five fresh frozen male torsos were tested at maximum internal rotation, neutral and maximum external rotation, at 0, 45 and 90 degrees of abduction Mihata 2020.

Tightening the posterior cuff, meaning infraspinatus and teres minor, decreased scapular upward rotation in maximum internal rotation at 45 and 90 degrees of abduction. Tightening the posterior capsule instead increased scapular internal rotation, and pulled the scapula into protraction in maximum external rotation at 90 degrees.

Muscle and capsule produced different scapular faults, and neither stayed where it was applied. This happened in a specimen with no nervous system in it, which fixes something important: the mechanical continuity is real on its own, and the neural control described above rides on top of it.

So the athlete arrives with a rotation deficit measured at the shoulder and a fault expressed at the shoulder blade. The elbow pays too, and The Elbow owns the valgus load that arrives there when the shoulder stops paying its share.

06Reading the failure pattern

Impingement, Instability and Labral Injury Overlap More Than They Separate

Most shoulder injuries in athletes trace to one of three patterns, impingement, instability or labral injury, and the patterns overlap more than they separate. Primary impingement is a structural narrowing under the acromion that compresses the cuff, classically in the older athlete with overhead pain that is worse at night.

Secondary impingement is the athlete version. Subtle anterior instability lets the head drift forward, which functionally narrows the space and pinches the cuff in a joint that looks structurally normal. The eighteen year old pitcher with anterior pain and the sixty five year old with night pain can have the same finding for opposite reasons.

The fatigue radiographs above make that concrete. A shoulder with no structural narrowing at all produced its own narrowing once the cuff ran out of capacity Chopp 2010. Space under the acromion is partly a regulated quantity, which is why the picture and the complaint come apart so often. Findings in People Without Pain carries the asymptomatic imaging tables that quantify how often.

The dead arm

Instability is the through line for the throwing athlete, and its signature complaint is the dead arm: sudden heaviness in the throwing shoulder and a loss of velocity that arrives without a tearing event. Rowe and Zarins put recurrent transient subluxation of the shoulder into the surgical literature in 1981 Rowe 1981.

The account that usually travels with the dead arm, that the humeral head tractions the brachial plexus during the subluxation, has no primary source this page could verify. It is named among the removals below. What the athlete reports is not in doubt, and neither is the timing: velocity goes before pain does.

Where the labrum sits in all of this

Labral tears, and the superior lesion at the biceps anchor above all, ride alongside instability and internal impingement rather than standing apart from them. That lesion is the SLAP tear, named for the superior labrum running from anterior to posterior. The collegiate surveillance data show what it costs.

SLAP tears carried a proportion ratio of 5.5 for recurring and 6.4 for running chronic Vargas 2022. Both figures describe a lesion that keeps coming back to the same athlete, in a sport where the mechanism that produced it goes on being performed.

The same dataset separates the two ways a shoulder gets hurt in one sport. Throwing was the mechanism in 87.5 percent of superior labral tears and in every case of rotator cuff tendinitis. Every anterior dislocation happened running the bases or sliding, and acromioclavicular sprains ran with contact at a proportion ratio of 20.6.

Reading which pattern drives a given shoulder is the entire diagnostic game, because the rehabilitation for each diverges sharply. The tests that separate them, with their sensitivity and specificity figures, belong to Orthopedic Screening, Upper Extremity.

07Examining the shoulder

The Shoulder Is Examined From the Neck Down Because Its Pain Refers

The shoulder is examined in a fixed sequence because its pain lies, and a clinician follows the order rather than chasing the symptom. Pain felt at the shoulder can originate in the cervical spine at C4 and C5, in the thoracic outlet, at the acromioclavicular joint, or in the heart, lung or gallbladder.

So the workup begins proximally, with a neurovascular and cervical screen, before the shoulder itself is touched. Then comes visual inspection, palpation, active and passive range, manual muscle testing, and only then the orthopedic cluster: impingement signs, acromioclavicular provocation, instability and apprehension testing, labral and biceps tests.

The neck earns its place at the front of that order on anatomy rather than on caution. The Cervical Spine in the Athlete carries the cervical spine under athletic load, and the spindle density behind its role as the sensory reference for where the arm is.

Weakness without pain is a neurological finding

A cuff that tests weak with no pain at all points toward a nerve rather than a tendon, and that distinction changes everything downstream. Ninety-six top-level volleyball players from eight teams were examined at the 1985 European Championships. Twelve of them had isolated paralysis of the infraspinatus on the dominant side, and none of them had symptoms Ferretti 1987.

Electromyography confirmed denervation. Isokinetic dynamometry put the strength loss on the affected arm at about 22 percent during external rotation. The investigators attributed it to repeated stretching of the suprascapular nerve during cocking and follow-through on the serve.

Hold what that means. Twelve athletes were competing at a European championship with a denervated external rotator and no complaint to bring to anyone. A pain-led examination finds none of them.

The pattern persists in the modern game at lower prevalence. Eighty-two professional volleyball players were examined after the jump topspin serve replaced the float serve. Infraspinatus hypotrophy, confirmed on magnetic resonance imaging, was present in 9 percent of the men and 12 percent of the women Mazza 2021. Every one of them had external rotation weakness, and none reported pain.

The long thoracic nerve carries the same logic at the other end. It drives serratus anterior, serratus anterior holds the scapula against the rib cage, and its failure shows as winging rather than as pain. The shoulder that will not stabilize is often a shoulder the brain is no longer reaching cleanly.

That is also why a shoulder examination in this practice tests drive and position sense alongside strength, and why imaging is ordered and interpreted rather than performed on site. Imaging the Athlete answers which image settles which question.

08A thrust away from the shoulder

A Thoracic Thrust Moved Scapular Rotation in One Trial and Left It Alone in Another

Two randomized trials asked whether an input delivered to the thoracic spine reaches the scapula, and they answered differently. In the first, 61 patients with shoulder impingement syndrome were randomized to thoracic spine manipulation or a sham version of it Haik 2017. Both the patient and the assessor were blinded, and the two sessions ran over one week.

Scapular upward rotation during arm lowering increased in the manipulation group by 5.3 degrees immediately after the second session and by 3.5 degrees at follow-up. Pain fell 0.7 points before the second session and 0.9 after it.

The same trial is equally clear about what did not move. Disability and rotator cuff index scores improved in both groups. Upper trapezius activity increased in the sham group. Middle trapezius, lower trapezius and serratus anterior activity decreased in both groups. The investigators state that manipulation did not appear to influence scapular muscle activity, and that the pain, function, scapular tilt and internal rotation results are not conclusive.

The trial that found nothing

The second trial found no effect at all. Fifty-two participants with subacromial impingement symptoms received a single session of thoracic manipulation or sham manipulation, and thoracic kinematics, thoracic excursion and scapular kinematics were measured before and after Kardouni 2015.

There were no significant between-group differences on any of them, or on pain, function and global rating of change. Both groups gained about 0.9 degrees of scapular internal rotation during arm raising and 0.8 during lowering, and both reported pain down 1.2 points.

One trial moved a scapular angle and one did not. That scatter is what the Unified Model of Tone predicts, because there is no such thing as an input acting upon an empty body. The same class of input meets a different starting organization in every subject, and the group mean is the average of those meetings rather than the effect of the input.

What both trials share is the question they thought worth asking. Neither research group measured the thoracic spine and expected the thoracic spine to answer. They measured the shoulder blade, because the scapula rides on the rib cage and lower trapezius runs to the thoracic spine. The load path is the assumption underneath the trial design.

09Rebuilding in stages

A Shoulder Returns to Overhead Load in Graded Stages, Not on a Calendar

Bringing a shoulder back to sport rebuilds control before it rebuilds load, and the sequence is not optional. The first stage calms the tissue and restores motion, working in the scapular plane and protecting the anterior capsule from the hyperextension that stresses the front of the joint. The second restores the force couples and the scapular rhythm.

The third layers strength and endurance into progressively sport-specific positions. The fourth returns the athlete to throwing or overhead load through a graded, criteria-based progression rather than a date. The criteria, the throwing programs and the scapular retraining evidence belong to Upper Extremity Rehabilitation.

Why everything happens in the scapular plane

The scapular plane is the plane the shoulder blade itself lies in, angled forward of the coronal plane, and it is where the research measures too. The fatigue radiographs were taken there Chopp 2010. The impingement kinematics were recorded during elevation there Ludewig 2000. Professional pitchers' rotation was measured with the arm abducted in it Wilk 2015.

That convergence is not a convention. It is the position in which the humeral head and the glenoid line up, so it is the position in which the cuff's compression acts along the socket rather than across it. Rehabilitation starts there for the same reason the laboratory measures there.

What threads every stage

The nervous system is the thread. Joint position sense at the shoulder is rebuilt deliberately, because the cuff cannot center a head whose position the brain cannot feel. Proprioception and Joint Position Sense carries the position sense measurements, including the throwing-shoulder findings and what fatigue does to repositioning accuracy.

The target is scapular control under fatigue rather than scapular control when fresh. Twenty-five overhead athletes with no history of shoulder pain showed altered lower trapezius activity and altered scapular upward rotation only after their external rotators were fatigued Joshi 2011. A rested shoulder answers a question the ninth inning never asks.

The thread runs wider than the joint. The model reads scapular precision as one local expression of the athlete's central integrative state, the level at which the whole system is running that week. A shoulder rebuilt in a rested athlete and the same shoulder in the fourth week of a heavy block are not the same shoulder. The staged progression is what keeps that difference visible.

So a shoulder does not get stable by getting strong. It gets stable by getting precise, and precision is a property of timing and position sense rather than of cross-sectional area. The goal is not a shoulder that has healed. It is a shoulder the brain trusts at full speed.

10What we corrected

Four Figures and a Quotation Removed From This Page

This page previously put the throwing arm's rotation at over seven thousand degrees per second. No primary source could be traced for that figure, so it is gone. What replaces it is the measured record: 67 newton meters of internal rotation torque, 380 newtons of anterior force during cocking, and 1,090 newtons of compression after release Fleisig 1995.

Second, the page stated that the glenoid covers about a quarter to a third of the humeral head, and carried 30 percent as a headline figure. No source supported it. The measured description of the socket replaces it, at roughly 9 millimeters of depth top to bottom and 5 front to back, with the labrum supplying about half Howell 1989.

Third, the page placed the scapular plane at roughly thirty to forty-five degrees forward of the coronal plane. The plane is real and the research measures in it, but that specific range could not be sourced and has been removed. Fourth, the account of the dead arm as the humeral head tractioning the brachial plexus could not be traced to a primary source either.

The page also carried a quotation attributed to Dr. Jason Dulberg that was not drawn from anything he said or wrote. It has been removed. Claims made here are either sourced to the literature or named explicitly as the model's.

11The model's claim

The Shoulder Is a Load Path That Loses Its Distribution

Two layers run through this page and they should not be confused. The established science is the socket anatomy, the concavity compression mechanism, the pitching kinetics, the injury surveillance, the dyskinesis prevalence and its prospective risk, the fatigue kinematics, the cadaveric tightness work and the manipulation trials. Each of those belongs to the investigators who measured it.

The Unified Model of Tone answers with one line: a local change in tension does not stay local. Applied to the shoulder, that changes what a shoulder injury is. The thing that fails is not a part. It is the distribution across the parts.

The scapula is the readout because it is the segment that has to answer everything else. The humerus hangs below it, the rib cage moves under it, the thoracic spine anchors the muscles that hold it, and the arm's whole load path runs through it on the way to the trunk.

The findings on this page fit that reading one after another. Tightness applied at the humerus moved the shoulder blade in a specimen with no nervous system Mihata 2020. Fatigue applied at the external rotators quieted the lower trapezius Joshi 2011. A painful shoulder used more trapezius in a worse position Ludewig 2000.

The prediction this page makes

Shoulder screening in overhead sport is dominated by rotation range: internal rotation deficit, total arc, flexion. Across 505 examinations of professional pitchers, none of those predicted shoulder injury or surgery Wilk 2015. The model says why. Each measures the size of an available range at one joint, and a shoulder does not fail by running out of range.

The measurement that would test the alternative fits inside a session a program already runs. Record scapular upward rotation at matched humeral elevation angles. Record scapular reposition error in degrees. Record RMSSD at rest. Record the time scapular kinematics take to return to their own baseline after a standardized overhead load.

The model predicts that set separates the athletes who go on to shoulder injury better than the rotation battery does, and it predicts it because all four are readings of one underlying organization. This is a claim about how a shoulder is regulated rather than a claim about what treatment does.

If scapular upward rotation, scapular reposition error, RMSSD, and time for scapular kinematics to return to baseline are shown to move together within the same athletes across a season, the unification claim is confirmed.

12The tone reading

The Scapula as the Readout

Three signatures of tone appear at the shoulder, each inside a measurement the literature already takes.

The rest of the library carries the same logic through its other foundations. Load is the 1,090 newtons of compression this joint absorbs after ball release, delivered into a socket 9 millimeters deep. Input quality is what a denervated infraspinatus stops reporting in an athlete who feels nothing wrong. Prediction is the feedforward model that positions the scapula before the arm moves rather than after it. Time course is the season across which external rotation grew roughly 10 degrees in both shoulders at once. Set point is the resting scapular position the periscapular muscles defend. Oscillation is the rhythm of the throw, which the shoulder blade has to enter and leave on time. The full framework is set out in the Unified Model of Tone.

13Where this sits

How This Page Relates to the Rest of the Library

Seven places this argument continues, each with the claim that earns the link.

14Questions athletes ask

Questions Athletes Ask

15The sources

References

18 primary sources, each linked to its record. Figures quoted on this page were checked against the published abstract.