Is this subscapularis reconstructable?

A 64 year old female presented with chronic right shoulder dysfunction and pain after an injury playing tennis in 2022, where she hit an overhead ball and felt a pop in her shoulder. She then had a recurrent injury in 2025 with a similar mechanism playing pickleball and was diagnosed with a rotator cuff tear. She had participated in PT and got some functionality back but remained weak, and then had a fall biking which further reduced the function of her shoulder. She complained of anterior shoulder pain and an inability to perform overhead activity due to weakness. Her exam revealed full active motion, but substantial weakness of belly press and lumbar lift-off. Passive external rotation was to 90 degrees. There was a palpable defect in the subscapularis. Plain radiographs were normal. MRI images are shown below.

In spite of our lack of optimism regarding the reconstructability of this chronic tear, the patient asked for an exploration and repair or reconstruction if possible.

At surgery the upper half of the subscapularis was detached of poor quality. With care to protect the axillary nerve, dissection of the lower half of the subscapularis was carried out  freeing it from the inferior capsule, and releasing the anterior capsule from the glenoid.

After this 360 degree release, excellent quality (> 1 cm thick) tendon with subjacent capsule could be easily approximated to the entire footprint at the lesser tuberosity. Six FiberWire sutures were passed through the lesser tuberosity and the tendon. The long head tendon of the bicep was incorporated in the repair. After securing the tendon, the shoulder externally rotated to 30 degrees with a firm endpoint. 

I thought you might find this interesting.

Building Back

Marsh Wren reconstructing nest

Montlake Fill

2010

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link)
The total shoulder arthroplasty (see this link)
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)

Shoulder rehabilitation exercises (see this link). 

Does subscapularis sparing = subscapularis sparing?

Clinically important subscapularis failure is reported in approximately 5% of patients following anatomic total shoulder arthroplasty. For example, Functional and radiographic results of anatomic total shoulder arthroplasty in the setting of subscapularis dysfunction: 5-year outcomes analysis found that patients who develop subscapularis dysfunction after TSA have worse patient reported outcome, range of motion, functional tasks of internal rotation, and radiographic outcomes, as well as increased rates of revision. In spite of subscapularis dysfunction, these patients maintained clinically significant improvement for pain and function at a mean 5-year follow-up.

Some surgeons have advocated a "subscapularis-sparing" approach to shoulder arthroplasty, with the goal of improving patient outcomes. The theoretical advantages of the subscapularis-sparing approach include earlier rehabilitation, reduced risk of subscapularis failure, and preservation of anterior shoulder stability.  See for example, The subscapularis-sparing windowed anterior technique for total shoulder arthroplasty which reviewed 47 cases performed by an individual experienced surgeon. 

However a recent article, Safety and Efficacy of Subscapularis-Sparing Shoulder Arthroplasty Approaches: A Systematic Literature Review considered 15 studies (1573 patients) reporting subscapularis-sparing shoulder arthroplasty. Eight studies directly compared subscapularis-sparing and standard techniques. The authors concluded "subscapularis-sparing" shoulder arthroplasty is a safe and effective technique with comparable patient outcomes, range of motion, pain and revision to nonsparing techniques." - i.e. they did not find evidence of greater clinical value with the "subscapularis-sparing" approach.

Two articles directly compared the two techniques (both by the same authors):

Total shoulder arthroplasty using a subscapularis-sparing approach: A radiographic analysis concluded "Although anatomic restoration of the shoulder can be accomplished using subscapularis-sparing TSA, retained osteophytes and significant mismatch of the HHD raise concerns regarding long-term outcomes."

Subscapularis-Sparing Total Shoulder Arthroplasty: A Prospective, Double-Blinded, Randomized Clinical Trial reported "At short-term follow-up, the outcome of TSA using the SSC-sparing surgical approach was similar to the outcome of TSA using the standard approach."

Concerns

(1) Like all new techniques, subscapularis-sparing has a learning curve to be negotiated by each surgeon using it. Even in experienced hands, in certain cases the technique may need to be abandoned in favor of a subscapularis takedown. See The subscapularis-sparing windowed anterior technique for total shoulder arthroplasty. 

(2) Risk of subscapularis distruption. Ultrasound assessment after a subscapularis-sparing approach to total shoulder arthroplasty found that rate of subscapularis disruption using a subscapularis-sparing approach for aTSA by was 14% (32 (86%) of 37 were intact)- thus, the risk of tendon injury is not eliminated with the subscapularis-sparing approach. As is the case for the failures with the standard approach, those with disrupted tendons had reasonable patient-reported outcomes but demonstrated decreased strength in forward flexion, abduction, and external rotation. 

(3) Limited exposure (See The subscapularis-sparing windowed anterior technique for total shoulder arthroplasty.)

    a. Restricted visualization may result in suboptimal component positioning, with 31.8% of patients showing center-of-rotation differences >3 mm. Additionally, humeral head diameter mismatch >4 mm occurs more commonly in the subscapularis-sparing group compared to traditional approaches.

    b. Incomplete osteophyte removal occurs more frequently with the subscapularis-sparing technique, achieved in only 75% of cases compared to more complete removal with traditional approaches. 

(4) Complications

    Complication profiles were similar between subscapularis-sparing and standard approaches in the randomized trial, with 3 patients in the sparing group and 2 in the standard group requiring revision surgery. See: Subscapularis-Sparing Total Shoulder Arthroplasty: A Prospective, Double-Blinded, Randomized Clinical Trial 

[2]Incomplete osteophyte removal occurs more frequently with the subscapularis-sparing technique, achieved in only 75% of cases compared to more complete removal with traditional approaches. 

Comment:

The subscapularis-sparing approach is being explored to minimize clinically significant subscapularis dysfunction after shoulder arthroplasty. 

We have not adopted the subscapularis-sparing approach at this point; instead we prioritize surgical exposure to optimize glenoid component seating and complete osteophyte resection. We have not conducted a head to head comparison of our method to the subscapularis-sparing technique.

We also recognize that all shoulders and all subscapularis tendons are not the same.  Thus, as the protocol below emphasizes (steps 1, 2 and 3), assessing in each patient the risk factors for subscapularis failure - regardless of the technique used - is essential to surgical decision making.

As Johnathan Swift (author of Gulliver's Travels) pointed out in 1750:

Our approach for anatomic arthroplasty is based on several key steps:

(1) Assess preoperative stiffness. Shoulders with limited external rotation when the arm is adducted are at higher risk of repair failure.

(2) Assess preoperative strength. Shoulders with weak internal rotation may have poor quality subscapularis tendons increasing risk of postoperative failure.

(3) Assess other risk factors for subscapularis failure: inflammatory arthropathy, malutrition, steroid use, prior surgery

(4) In shoulders at high risk for subscapularis failure, consider a reverse rather than an anatomic total shoulder (aTSA).

(5) For patient having aTSA , release subscapularis tendon and subjacent capsule completely from lesser tuberosity, retaining capsule on the tendon's deep surface.

(6) Perform a 360 degree release of the subscapularis from the coracoid, glenoid, and inferior capsule to optimize excursion of tendon

(7) Perform glenoid arthroplasty

(8) If necessary, trial undersized humeral head component so that the lateral border of the subscapularis reaches the reattachment site at the lesser tuberosity with the arm in external rotation.

(9) Pass six FiberWire sutures through quality bone at the lesser tuberosity

(10) Insert humeral component sized per trialing (#5 above)

(11) Place additional FiberWire sutures in the rotator interval to reinforce the repair.

(12) Tie repair sutures

(13) Verify satisfactory motion before skin closure.

(14) Start assisted flexion in recovery room (note: our practice is to avoid plexus blocks for shoulder arthroplasty to allow sensory feedback during these exercises).

(15) Allow use of arm for activities with elbow at side. Gradually transition to active elevation starting at six weeks after surgery.

(16) Allow progressive increase in resistance over the first postoperaive year, making sure that 20 repetitions are easy and comfortable before adding additional weight.

Conclusion

Current evidence suggests that standard and subscapularis-sparing approaches produce similar outcomes. We prefer subscapularis peel and secure repair because in our hands it provides excellent exposure for osteophyte resection and glenoid component positioning and seating. At the same time we recognize that other surgeons may be equally comfortable with the subscapularis-sparing approach.

Making Choices

Cliff Swallows

Kalaloch Lodge, Washington

2021

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link)
The total shoulder arthroplasty (see this link)
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)

Shoulder rehabilitation exercises (see this link). 

Head splitting fracture, ORIF, post traumatic arthritis, ream and run, subscapularis failure, graft reconstruction

 A 51 year old active man sustained a head splitting fracture in a high velocity ski injury.

This was treated at an outside hospital with open reduction and plate fixation. Radiographs showed an incomplete reduction with posterior displacement of the proximal humerus and posterior decentering of the head on the glenoid,

By six months after surgery, the shoulder was stiff, painful, and arthritic.

He presented to us requesting a ream and run procedure with a chrome cobalt humeral head. At the time of this surgery his plate was preserved to minimize the risk of postoperative fracture. The head was centered using soft tissue balancing.

He did well for the the first 2 months after surgery, when he started himself on a vigorous external rotation stretching program. This led to abrupt failure of his subscapularis.

After discussion of the alternatives of non-operative management, reverse total shoulder or attempted subscapularis reconstruction with allograft, he elected the latter.

While his procedure was on the right shoulder, our technique of hamstring reconstruction is shown below for a left shoulder.

In this procedure the retracted subscapularis muscle is carefully freed from the coracoid muscles, neuromuscular structures, and glenoid.

Two holes are made in the lesser tuberosity using a pinecone bur. These holes are connected using curved curettes.

A high quality, long hamstring tendon graft is prepared and threaded through these holes.

Two vertical slits are made in quality subscapularis tissue

The ends of the hamstring graft are passed through these slits.

The graft strands are pulled laterally, approximating the subscapularis to the lesser tuberosity.

And secured to the bone lateral to the lesser tuberosity.

This technique provides four strong strands of allograft that can reinforce a tendon repair or span a gap if the mobilized tendon cannot reach the lesser tuberosity without undue tension.

The mobility and the security of the repair is verified.

Two years after his repair, the patient reported he was doing well, being active in Tai Chi, swimming, weight lifting and throwing with his son, having only minimal discomfort. On examination he had 160 degrees of active flexion with good strength, external rotation to 45 degrees and 4/5 strength of belly press.

His x-rays show a stable, centered humeral head and a remodeling, reamed glenoid. 

This procedure is a valuable procedure for those young, active patients with a failed subscapularis after shoulder arthroplasty who wish to avoid a reverse total shoulder.

Considering the big picture

Mt. Rainier

July 4, 2025

You can support cutting edge shoulder research that is leading to better care for patients with shoulder problems, click on this link

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).

 

Learning from surgical failure

 On 22 October 1895, there was a particularly spectacular railway accident at the Gare de l'Ouest in Paris. The No.56 train arriving from Grandville hurtled into the station at a speed of 40 to 60 kilometres an hour, and, unable to stop, plowed through the buffers at the end of the platform. Its engine crashed through the façade of the station building, and fell down on to the Place de Rennes.

Hopefully the responsible railway system, Chemins de fer de l'Ouest, learned more from studying this failure than from studying all their on time, safe arrivals.

The May 26, 2025 post, "Do higher case volumes make us safer and more effective surgeons?", pointed out that it is not the number of cases we do that make us better, but rather the study of our failures that provide a path to improvement. This post also pointed out that the surgeon performing the procedure is, in many instances, the most important determinant of the outcome: the surgeon is the method. 

I found the recent article: Anatomic Total Shoulder Arthroplasty using a Short Humeral Stem and a Non-Augmented Minimally Cemented All-Polyethylene Glenoid: Minimum 2-Years Outcome and Predictors of Clinical Failure of interest in this regard because - in contrast to most other clinical reports -  it presents the experience of an individual senior shoulder surgeon. In the words of Kahneman, this filters out the pattern noise when multiple surgeons are included in the report (see the striking chart in the May 26 post).

The authors reported excellent average outcomes for 128 consecutive anatomic total shoulder arthroplasty (aTSA) using non-augmented glenoid implants and short humeral stems for glenohumeral arthritis with an intact rotator cuff. In spite of the average success, they sought to identify factors associated with adverse outcomes.

The surgeon's technique included minimal glenoid reaming to achieve >90% backside contact of a pegged, all polyethylene standard glenoid implant, inserted with minimal cement. The surgeon used a subscapularis peel which was repaired after insertion of the components using 6 to10 high-strength #2 braided sutures passed through transosseous drill holes. After the implants were placed, the subscapularis tendon was closed with these sutures and with closure of the distal rotator cuff interval with additional braided sutures to reinforce the repair.

104 of the 128 aTSAs were available for evaluation at minimum 2 years follow-up (range 2-5.6 years) (see the challenge of longer term followup). In comparison to many published reports, this represents a high percent followup. The authors included information on the remaining 24 - a few had passed on or became invalid, a few were revised and only a dozen or so could not be tracked. In clinical studies the denominator counts big and the reasons for loss to followup are important.

As stated in A Study in Hospital Efficiency (Codman, E.A. 1917). "All patients should be followed long enough to determine whether or not the treatment has been successful, and to ask 'if not, why not?'".

Overall, mean ASES scores improved from 41 to 89, SST from 4.5 to 10, and VAS-pain from 5.5 to 0.7. [Note that these average postoperative scores are within the minimal clinically important differences of perfect scores.  Thus it would be statistically impossible for any change in implant or technique (for example, routine use of preoperative 3D CT planning) to make a clinically significant improvement in the average outcomes.]

Fifty-nine patients had preoperative posterior glenoid wear patterns (Walch B2 or B3) and 47 had concentric or minimal glenoid wear (Walch A1, A2, or B1). The presence of posterior glenoid wear was not found to influence any outcome score

There were three modes of failure identified by the authors

(1) Subscapularis failure: 5 patients underwent revision to reverse shoulder arthroplasty for subscapularis insufficiency and one underwent open subscapularis repair with graft augmentation following traumatic injury. Three of the subscapularis tears arose because of trauma and/or patient reported noncompliance with postoperative immobilization or restrictions. All but one of the subscapularis failures reported here occurred within the first 12 months postoperatively. None of the patients with subscapularis failure requiring surgery had undergone prior open capsulorrhaphy or subscapularis repair

(2) ASES score <70: Ten patients had a final ASES score < 70, which was associated with a history of previous surgery (overall, twenty-four shoulders had undergone previous surgery). 

(3) Glenoid osteolysis: 10 patients (13%) had radiographs  demonstrating glenoid osteolysis (defined as "radiolucency extending at least two mm from the center of peripheral pegs"). Glenoid osteolysis was not found to be associated with preoperative posterior glenoid erosion. The presence of glenoid osteolysis was not found to have an adverse effect on clinical outcome with the period of followup in this study. 

Comment: These authors report overall excellent results with their surgical technique. They have also defined three adverse outcomes that merit further study so that they and we can learn to avoid them.

Subscapularis failure: It would be of interest to know the rehabilitation program used by the surgeon. I am using a "slow roll" approach to range of motion exercises during the first six weeks and spend time cautioning the patient about falls and about avoiding external rotation stretching and active internal rotation during this period. Perhaps this surgeon does the same.

Prior surgery: It's a fact that young patients with arthritis commonly have had prior surgery. While the patient's history is not modifiable, we should try to learn if these patients have particular issues that could be addressed at surgery, such as (a) stiffness that might drive consideration of more aggressive soft tissue releases or smaller components or (b) difficulty in centering the humeral head on the glenoid that may drive use of eccentric humeral head components.

Glenoid osteolysis: While glenoid osteolysis was not noted to have an adverse effect on clinical outcomes in this minimum two year followup study, longer term followup is needed, although such studies are challenging.  These authors took care to optimize glenoid bone preparation. It would be of interest to know how well the components were seated, in that glenoid seating may be a major factor in achieving durable fixation.

This is a fine example of an individual surgeon study that has identified three adverse outcomes from anatomic arthroplasty. While it is unlikely that modifications in the surgical technique will lead to clinically significant improvement in the average outcome scores (because the average scores reported are so good), it is possible that additional study of patients with these failure modes may show us how to lower their frequency. 

The first step is to define the problems.

Fish with a problem

Montake Fill, Seattle June 2019

You can support cutting edge shoulder research that is leading to better care for patients with shoulder problems, click on this link

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).

1. 
   

   185. 185. 
             

        186. 185. 
                  

                  208. 
                       

                  209. 
                       

                       231. 
                            

                       232. 300. 
                                 

                            301. 
                                 

                            302. 
                                 

                                 300. 346.