Rotator cuff repair results in predictable pain relief and a variable return of function in the majority of patients. However, it may take up to a year to achieve maximum improvement in shoulder strength and function. In certain cases patients may continue to experience pain and weakness in the shoulder following rotator cuff repair. Symptomatic failed rotator cuff repair continues to be a challenge for an orthopaedic surgeon. However, not all structural failures (as seen on imaging) are clinically symptomatic, and patients who don’t fully heal their rotator cuff may continue to perform at a higher level of function and achieve considerable pain relief. Therefore, the term “rotator cuff failure” includes presence of significant pain, stiffness, or weakness, or a combination of these three presentations irrespective of the radiographic findings.
There are three critical questions that the orthopaedic surgeon should ask before proceeding with the revision rotator cuff surgery:
- Why did the first surgery fail?
- How does one go about fixing the problem this time?
- Are there any pre-operative risk factors present in the index patient that are predictive of a high rate of failure with the revision surgery?
Patients with rotator cuff failure can present with shoulder pain and/or significant weakness or stiffness. The post-operative shoulder pain can be at a new anatomic location or in a similar location to the pre-operative pain. A patient presenting with post-operative shoulder pain that is similar in location and character to the pre-operative pain can present as one of the three patterns:
- Shoulder pain that never improved after surgery (wrong diagnosis with alternate etiology for the shoulder pain, peri-operative rotator cuff failure, post-operative adhesive capsulitis).
- Interval improvement in the symptoms after surgery but acute onset of shoulder pain following a traumatic event (acute retear of rotator cuff).
- Interval improvement in the symptoms after surgery followed by insidious recurrence of similar pain in the index shoulder (chronic recurrent tear).
Patients may also present with loss of strength and functional limitations independent of the shoulder pain. Loss of motion or stiffness can present early or late after rotator cuff repair.
Assuming that the previous rotator cuff repair was performed for a correct diagnosis and other concomitant diagnoses or sources of shoulder pain were addressed in the previous surgeries, there are multiple factors that can lead to failure of rotator cuff repair. See Table I.
A systematic approach should be utilized to investigate other causes (besides rotator cuff failure) that can result in shoulder pain and loss of function after rotator cuff repair (See Table II, Figure 1, Figure 2, Figure 3). These could be related to the cervical spine, biceps tendon, acromioclavicular (AC) joint, suprascapular nerve, adhesive capsulitis, and infection. All previous patient records (office visits, operative notes) and prior imaging studies must be obtained and reviewed.
Classic Physical Exam Findings
A shoulder examination starts with examination of the cervical spine to rule out cervical radiculopathy, which could be a co-existing pathology. Inspection of the shoulder may demonstrate atrophy of the rotator cuff muscles in the supraspinatus fossa and infraspinatus fossa (Figure 4), partial or complete deltoid atrophy (especially with previous open or mini-open cuff repair), anterosuperior prominence of the humeral head, and scapular dyskinesia.
The patient should be asked to point to the site of maximum pain before proceeding with palpation. Location of the pain and the site of maximum tenderness are valuable in the anatomic localization of the pain generating structure. Shoulder range of motion (ROM) (active and passive) is usually tested next. The differences in the internal and external rotation of the shoulder in adduction and horizontal abduction provide important information about the anatomic location of capsular contracture. The resisted strength testing of the rotator cuff muscles helps in detecting the weakness of the subscapularis, supraspinatus, infraspinatus, and teres minor (empty can test, belly press test, resistance in external rotation). It can be challenging to perform resisted strength testing of rotator cuff in a failed rotator cuff surgery especially in the presence of significant pain and stiffness.
Special tests for impingement (Hawkins and Neer’s sign), massive cuff tears (external rotation lag sign, horn blower sign), AC joint pathology (cross arm adduction, O’Brien’s test), and biceps pathology (Speeds, Yergason, O’Brien’s test, subpectoral biceps test) should be performed. A complete neurovascular examination including tests for the evaluation of thoracic outlet syndrome should be performed.
The imaging studies are generally helpful in identifying a structural rotator cuff defect, and identifying alternate pathology (missed pathology) that could explain residual shoulder symptoms. Imaging is also important in assessing the status of the residual rotator cuff (atrophy, retraction, and fatty infiltration; Figure 5, Figure 6), AC joint, and biceps tendon.
Plain radiographs of the glenohumeral joint including the anteroposterior (AP) view in internal and external rotation, axillary, and supraspinatus outlet views should be obtained (Figure 6, Figure 7, Figure 8). The presence of degenerative changes in the glenohumeral joint, presence of associated pathology (os acromiale, AC joint arthrosis, and calcific tendinitis), and position of the previous hardware (metallic anchors) can be seen on a plain film. The AP view in internal and external rotation is useful in assessing bone loss or osteolysis in the greater tuberosity following previous rotator cuff repair.
Advanced imaging studies are helpful in demonstrating a rotator cuff defect but whether this defect actually correlates with pain and loss of function requires clinical correlation. Magnetic resonance arthrography (MRA), computed tomography arthrography (CTR), and ultrasonography (USG) are helpful and have their respective indications. The MRA is considered the imaging of choice for rotator cuff assessment in the revision surgery. The USG is cost beneficial but is highly user dependent. The CTR is a valid option in patients who have a contraindication to magnetic resonance imaging (MRI) (implantable devices, pacemakers). However, there are limitations to advanced imaging modalities as they may potentially either undercall or overcall the level of healing of the tendon. Furthermore, presence of previous hardware can result in artifacts, which can further limit the interpretation of imaging studies (Figure 6).
Non-operative management is directed towards reducing inflammation and regaining shoulder ROM and strength. Non-operative treatment options include anti-inflammatory medications (non-steroidal anti-inflammatory agents) and corticosteroid injection. Activity modifications, treatment modalities (heat therapy, cold therapy, USG), and physical therapy are helpful in regaining strength and motion.
Following a rotator cuff repair, some patients can demonstrate a prolonged lag phase before they improve clinically in terms of pain relief and return of motion and strength. It may take up to a year before maximum recovery of strength is achieved. Patient counseling and emotional support plays a strong role in getting these patients back to the path of favorable recovery.
Indications for Surgery
Patients who have failed a course of non-operative management and continue to be in significant pain, and are stiff and weak are potential candidates for revision surgery. However, advanced age, advanced fatty infiltration of the rotator cuff (Goutallier stage >2), chronic massive cuff tears, and severe cuff tear arthropathy are relative contraindications to the revision rotator cuff repair. The aforementioned conditions are predictive of poor healing of the rotator cuff and will result in high failure rate of revision repair. The final determination of the feasibility of complete versus partial repair is made intra-operatively based on multiple factors including the quality of the rotator cuff tendon, size of the defect, and mobility of the rotator cuff.
Pre-operatively, the biceps and AC joint are evaluated to make sure that they are not symptomatic. If the history and clinical examination does not provide a clear distinction, a diagnostic injection of local anesthetic can be utlilized. It is very important to discuss the realistic expectations with the patient regarding outcomes from revision rotator cuff repair. Patients should be counseled and encouraged to quit smoking if they are current smokers. Consent should include use of allograft or synthetic rotator cuff patches to address large cuff defects. Biologic augmentation of the rotator cuff repair with autologous bone marrow aspirates and autologous platelet rich plasma (PRP) should also be discussed pre-operatively.
Previous operative notes should always be obtained if possible as they provide valuable information about the quality of rotator cuff tissue during previous surgery, the type of suture anchors used, the type of repair performed (single versus double), and if acromioplasty was performed during prior surgeries.
See also Figure 9:
- Basic shoulder arthroscopy set (probe, scissors, grasper, switching sticks, tissue punch, suture retriever, crochet hook, knot pusher, knot cutter).
- 30 degree and 70 degree arthroscope (4 mm).
- Articulated arm holding device.
- Arthroscopic shaver and burr.
- Arthroscopic cannulas.
- Arthroscopic radiofrequency ablation device.
- Suture shuttling/passing devices.
- Suture anchors (PEEK [our preference], metal, biocomposite, swivel locks, push locks).
- Sutures (FibreWire, Fibretape, polydioxanone (PDS) suture).
- Beach chair or lateral decubitus positioning devices.
- Biceps tenodesis implants (interference screw, anchor, or endobutton devices).
- Platelet rich plasma harvest kit, bone marrow aspiration kit.
Pre-operative intravenous antibiotic prior to the start of the procedure (within an hour).
Peri-operative deep venous thrombosis prophylaxis
Pneumatic compression devices on both lower extremities.
We prefer general endotracheal tube anesthesia and interscalene regional anesthesia for all our revision rotator cuff surgeries.
A beach chair position or lateral decubitus position can be used. We prefer a beach chair position for our rotator cuff repairs (Figure 10). The patient is brought to the edge of the table and safely secured to table. Two folded towels are placed behind the scapula and the contralateral arm is supported on a hand rest or pillow. The head and neck are safely secured to the table and excessive flexion or extension of the neck is avoided. A heating blanket is then placed on the lower part of the torso and legs.
The operative time for revision rotator cuff repair can be longer and there is always a concern for swelling of the shoulder. We prefer our pump pressures to be between 40 and 50 mm Hg and work closely with the anesthesiologist to keep the systolic blood pressures appropriate (usually ~100 mm Hg) for adequate visualization throughout the surgery.
Arthroscopic revision rotator cuff repair: Surgical steps
- Exam under anesthesia (EUA): EUA provides valuable information regarding shoulder ROM and pattern of capsular contracture and is performed prior to diagnostic arthroscopy.
- Arthroscopy portals and diagnostic arthroscopy: Old portal sites can be used if they are in the right location. A thorough diagnostic shoulder arthroscopy is performed through a standard posterior portal. A standard anterior portal is established in the rotator interval immediately lateral to the coracoid after localization with a spinal needle. Close attention should be paid to the long head of the biceps (LHB) insertion and the intra-articular and intertubercular portion of the biceps tendon. We have a low threshold to perform tenotomy during revision surgery if there are any signs suggestive of tendonitis or significant tear. Subscapularis tendon tears are less common compared to the posterosuperior cuff tears but they are often missed or misdiagnosed as just fraying of the upper rolled edge. The medial and lateral part of the subscapularis footprint should be meticulously examined and this can be accomplished with the arm in flexion and internal rotation. The supraspinatus, infraspinatus, and teres minor tendons are examined for the extent of tendon tear, degree of tendon retraction, quality of rotator cuff tissue, and presence of delamination. Chondral wear or damage on the humeral head or glenoid should be documented. Loose suture material or implants are removed (Figure 11, Figure 12). If there is any sign suggestive of infection, multiple synovial and rotator cuff biopsy specimens should be obtained and sent for culture and sensitivities.
- Intra-articular work: After the diagnostic glenohumeral arthroscopy is completed we proceed with the intra-articular work, which includes:
Biceps tenotomy: if the LHB tendon demonstrates significant tear, instability, intense synovitis, or degeneration we will perform a tenotomy (Figure 13, Figure 14, Figure 15). The decision to perform biceps tenodesis of the tenotomized tendon usually takes into account multiple factors and includes the patient’s age, occupation, level of activity, and perception of the popeye deformity. However, mini open subpectoral biceps tenodesis is our preferred technique for LHB tenodesis but it can be accomplished arthroscopically as well at the suprapectoral location.
Debridement of the labral fraying or tear.
Capsular release: A complete capsular release is performed if there is post-operative capsular contracture from prior surgery. The posterior capsular release is performed from the posterior working portal and anterior viewing portal (Figure 16, Figure 17, Figure 18).
Fraying or minimal tears of the upper border are treated with debridement alone. If there is significant fraying of the upper border but the footprint is intact, we perform a side-to-side repair using a delayed absorbable suture. If a subscapularis is torn off the footprint we will repair the tendon unless the tear is really small, where debridement alone is performed. We prefer arthroscopic subscapularis tendon repair unless the tear is massive and chronically retracted.
An anterolateral working portal is established approximately 2 finger breadth off the edge of anterolateral corner of the acromion. The anterolateral portal and anterior portal constitute the working portals and the posterior portal is used as the viewing portal. The basic principles of arthroscopic rotator cuff repair should be followed during revision repair as well and includes tendon mobilization, tear pattern recognition, footprint preparation, and performing a tension free repair. Subscapularis repair is performed sooner during the revision surgery as shoulder swelling can limit visualization and working in this tight space if performed later during the surgery.
Subscapularis tendon mobilization: The rotator interval is opened up via the anterior portal. Coracoidplasty can be performed from the anterior portal at this stage (Figure 19). The subscapularis tendon is mobilized by releasing the adhesions on the anterior, superior, and posterior surface of the subscapularis. Great care should be taken not to debride the tendon accidentally especially in chronic retracted tears where the tendon morphology is distorted. Also, the lateral part of the coracohumeral ligament adjacent to the supraspinatus should be preserved as it provides a visual guide for correct reattachment relationship between the anterior and superior rotator cuff during repair. An arthroscopic grasper introduced from the anterolateral portal can be used to pull the retracted tissue, which helps to define the distorted anatomy of the subscapularis in chronic retracted tears. A traction stitch can also be used during mobilization of the tendon.
Reducibility of the tendon: The next step is to assess the reducibility of the tendon to the footprint on the lesser tuberosity using an arthroscopic grasper from the anterolateral portal.
If there is good quality tendon tissue without significant retraction, we prefer a double row repair.
If there is significant retraction and tendon tissue loss, we proceed with a tension free single row repair.
Preparation of the footprint: This can be done via the anterior or the anterolateral portal using a shaver, burr, and radiofrequency probe. The cortical surface is freshened down to the bleeding bone. Bone marrow stimulation can be performed prior to use of suture anchors.
The anterior portal is used for inserting the suture anchors in the bone and shuttling sutures through the subscapularis tendon.
The anterolateral portal is used to tie arthroscopic knots (Figure 19, Figure 20, Figure 21).
- Subacromial work
The arthroscope is introduced into the subacromial space and a lateral portal is established under arthroscopic vision. Subacromial debridement is performed in the subacromial space, and the anterior, lateral, and posterior gutters releasing adhesions between the deltoid and rotator cuff. Sometimes, adhesions can be present between the rotator cuff and undersurface of the acromion, which should be recognized prior to starting the subacromial bursectomy. We usually perform acromioplasty at this stage. However, we follow a very minimalist approach with the acromioplasty unless there is a big acromion spur (Figure 22).
- Superior and posterior rotator cuff repair
The basic principles of arthroscopic rotator cuff repair should be followed during revision posterosuperior rotator cuff repair and includes tendon mobilization, tear pattern recognition, footprint preparation and performing a tension free repair.
Rotator cuff mobilization: The torn posterosuperior rotator cuff is mobilized first and freed from any adhesions to the surrounding deltoid and subacromial tissue. A traction stitch can also be used during mobilization of the tendon (Figure 23). Delamination and tendon tissue loss (intra-substance tendon tear) can result in significant rotator cuff defects. Tendon tissue loss can be assessed by estimating the amount of tendon still attached to the footprint and also from amount of tendon tissue lateral to the musculotendinous junction. After this, the reducibility of the torn rotator cuff to the footprint is assessed. If there is not enough reducibility of the tendon, soft tissue releases are performed. Subcapsular release above the glenoid is usually performed intra-articularly between the cuff and glenoid. One must be careful with the medial extent of this subcapsular release above the glenoid or close to the spine so as not to damage the suprascapular nerve. The anterior and posterior interval slides provide additional mobilization of the rotator cuff.
If the rotator cuff is deemed reparable down to the footprint, the greater tuberosity is prepared. The cortex is debrided of any residual soft tissue down to bleeding bone (Figure 23, Figure 24) and bone marrow stimulation is performed with a power pick or microfracture awl (Figure 25). Great care should be taken not to violate the cortical bone completely because it can lead to compromised purchase of the suture anchor especially in osteoporotic bone.
We prefer transosseous equivalent double row rotator cuff repair if there is minimal tendon tissue loss and the tendon can be brought back to the footprint without any significant tension (Figure 26).
Single row rotator cuff repair is performed if there is some tendon tissue loss and the torn tendon can be brought to the medial part of the footprint (Figure 27)
Tendon tissue loss or tendon defects are challenging situations (Figure 23, Figure 28, Figure 29). If the tendons cannot be brought to the footprint in their entirety despite adequate mobilization and soft tissue releases, partial repair of the torn tendon can be performed. This includes the use of margin convergence (side-to-side repair) and repair of the reducible component of the cuff to the footprint (Figure 30). Allograft patch (synthetic or biologic) augmentation is used for reinforcement of the tendon defects (Figure 31). The allograft can be biologically augmented with autologous platelet rich plasma (harvested from peripheral blood) and autologous bone marrow aspirate (harvested from the proximal humerus) (Figure 32).
- Tendon transfers (Latissimus dorsi transfer) and arthroplasty (reverse total shoulder arthroplasty) are alternative treatment options for an irreparable revision rotator cuff tear (Figure 29).
Pearls and Pitfalls of Technique
- Identify other sources of pain in the setting of failed rotator cuff surgery (biceps tendon, AC joint, glenohumeral arthritis, adhesive capsulitis, cervical radiculopathy).
- Pre-operatively identify irreparable rotator cuff tears and presence of risk factors for failure of rotator cuff healing (massive tears, severe fatty infiltration, advanced physiologic age) as these patients may not benefit significantly from revision rotator cuff repair.
- Use of delayed post-operative rehabilitation protocol for revision surgery for massive and large rotator cuff repairs.
- Failure to recognize significant tears of the subscapularis tendon.
- Failure to adequately mobilize the rotator cuff tendons during surgery.
Complications of Revision rotator cuff repair
- Failure of the rotator cuff tear to heal.
- Hardware related complications: greater tuberosity fracture, loose body, dislodgement, hypersensitivity reaction to suture anchors.
- Adhesive capsulitis.
- Infection (use of allograft patch).
- Deltoid insufficiency (open repairs).
Post-operative rehabilitation is an integral component of revision rotator cuff repair. We prefer a delayed rehabilitation protocol for revision rotator cuff repairs. The rationale behind a delayed rehabilitation program is that repair site micro motion and gap formation may negatively affect tendon healing. We immobilize our patients in a sling for 6 weeks to give the repaired tendon a chance to heal. The phases of rehabilitation are similar to the traditional method of rehabilitation following rotator cuff repair except that they do not start until 6 weeks after repair. The different phases of rehabilitation for revision rotator cuff repair can be summarized into following:
- Phase I (0-6 weeks): Immobilization in a sling with no passive range of motion (PROM) of shoulder. Active ROM of elbow, wrist, and fingers and shoulder pendulums are allowed.
- Phase II (6-14 weeks): The shoulder brace is discontinued at the start of this phase. Therapist supervised PROM of shoulder and scapula is started. Active assisted and active ROM exercises are gradually introduced after full passive ROM is achieved.
- Phase III (14-24 weeks): Strengthening exercises of the shoulder are introduced starting with isometric exercises, resisted bands, and gradually progressing to light weight exercises. Sports-related rehabilitation and advanced conditioning is initiated after 18-20 weeks.
- Phase IV (24 weeks onwards): Unrestricted return to daily activities.
Outcomes/Evidence in the Literature
The short term and medium term outcome studies demonstrate that the retear rates are higher and objective improvement in strength and functional outcome scores are lower and less predictable with the revision rotator cuff repair compared to the primary rotator cuff repair. Interestingly, multiple studies have demonstrated that revision rotator cuff does result in reliable pain relief in most of the patients.
Keener, JD, Wei, AS, Kim, HM, Paxton, ES, Teefey, SA, Galatz, LM, Yamaguchi, K. “Revision arthroscopic rotator cuff repair: repair integrity and clinical outcome”. J Bone Joint Surg Am. vol. 92. 2010. pp. 590-8. (Level 4 evidence; arthroscopic revision rotator cuff repair results in reliable pain relief and improvement in shoulder function in selected cases. Approximately half of the revision repairs were intact at 1 year post-operatively. Patient age and number of torn tendons significantly influence post-operative tendon integrity.)
Piasecki, DP, Verma, NN, Nho, SJ, Bhatia, S, Boniquit, N, Cole, BJ, Nicholson, GP, Romeo, AA. “Outcomes after arthroscopic revision rotator cuff repair”. Am J Sports Med. vol. 38. 2010. pp. 40-6. (Level 4 evidence; arthroscopic revision rotator cuff repair results in improved pain relief and shoulder function. Patients with more than one ipsilateral shoulder surgery and female gender are associated with poor results.)
Le, BT, Wu, XL, Lam, PH, Murrell, GA. “Factors predicting rotator cuff retears: an analysis of 1000 consecutive rotator cuff repairs”. Am J Sports Med. vol. 42. 2014. pp. 1134-42. (Level 3 study; retrospective evaluation of 1000 consecutive arthroscopic rotator cuff repairs for retear rates and risk factors for retear. Retear process is multifactorial and tear size including tear thickness, tear size area, and tear dimensions have a strong association with the retear rate.)
Jost, B, Zumstein, M, Pfirrmann, CW, Gerber, C. “Long-term outcome after structural failure of rotator cuff repairs”. J Bone Joint Surg Am. vol. 88. 2006. pp. 472-9. (Level 4 study; long term results (average 7.6) in 20 patients that had a failed rotator cuff repair and continued to do better with respect to pain and outcome scores compared to their presurgery status.)
Hartzler, RU, Sperling, JW, Schleck, CD, Cofield, RH. “Clinical and radiographic factors influencing the results of revision rotator cuff repair”. Int J Shoulder Surg. vol. 7. 2013. pp. 41-5. (Good results after rotator cuff repair associated with male gender, larger acromiohumeral interval, absence of shoulder arthritis, presence of preserved pre-opeartive glenohumeral motion, and absence of degenerative tear.)
Dodson, CC, Kitay, A, Verma, NN, Adler, RS, Nguyen, J, Cordasco, FA, Altchek, DW. “The long-term outcome of recurrent defects after rotator cuff repair”. Am J Sports Med. vol. 38. 2010. pp. 35-9. (Level 4 study; patients with failed rotator cuff repair and persistent rotator cuff defect can remain asymptomatic for a long term period of time but they lose strength and their tears continue to increase in size.)
Lo, IK, Burkhart, SS. “Arthroscopic revision of failed rotator cuff repairs: technique and results”. Arthroscopy. vol. 20. 2004. pp. 250-67. (Level 4 study; elegant technique description for revision rotator cuff repair.)
Ladermann, A, Denard, PJ, Burkhart, SS. “Revision arthroscopic rotator cuff repair: systematic review and author’s preferred surgical technique”. Arthroscopy. vol. 28. 2012. pp. 1160-9. (Level 4 systematic review; arthroscopic revision rotator cuff repair frequently leads to improved shoulder pain and function. Poor results associated with female gender, post-operative structural failure, poor pre-operative shoulder ROM.)
Djurasovic, M, Marra, G, Arroyo, JS, Pollock, RG, Flatow, EL, Bigliani, LU. “Revision rotator cuff repair: factors influencing results”. J Bone Joint Surg Am. vol. 83-A. 2001. pp. 1849-55. (Results of revision rotator cuff repair are inferior to the primary rotator cuff repair. Good results are associated with presence of intact deltoid, good quality rotator cuff tissue, preserved pre-operative ROM, and one prior shoulder surgery.)
Failed rotator cuff surgery is a challenging clinical situation for an orthopaedic surgeon. Symptomatic rotator cuff tendon retears, stiffness, and presence of significant weakness can result in dissatisfaction after rotator cuff repair. Identification of factors that led to failure of the index procedure, and a systematic approach to the management of failed rotator cuff surgery is essential for a successful outcome. Patients should be evaluated for other causes of shoulder pain after rotator cuff repair (biceps tendinosis, AC joint arthritis, adhesive capsulitis, and cervical radiculopathy). Chronic massive rotator cuff tears, rotator cuff tear arthropathy, multiple tendon tears (>2 tendons) and advanced fatty infiltration of rotator cuff (Goutallier stage >2) are scenarios that are less likely to have a successful repair and alternative treatment options should be considered.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- The Problem
- Clinical Presentation
- Diagnostic Workup
- Non–Operative Management
- Indications for Surgery
- Surgical Technique
- Pearls and Pitfalls of Technique
- Potential Complications
- Post–operative Rehabilitation
- Outcomes/Evidence in the Literature