Robert Z. Tashjian, Jay D. Keener, and Ken Yamaguchi
DEFINITION
Rotator cuff disease encompasses a spectrum of disorders ranging from tendinitis to partial and full-thickness tendon tearing.
It is the most common shoulder disorder treated by an orthopedic surgeon, with over 17 million U.S. individuals at risk for the disabilities caused by the disease.
The prevalence of full-thickness tearing of the rotator cuff ranges from 7% to 40% across multiple studies.17,20
Age-related degenerative change is a primary factor in the development of rotator cuff tears.24
Asymptomatic full-thickness tears have been found in 13% of the population between age 50 and 59 and in over 50% of people older than 80 years old.20
The risks and benefits of both nonoperative and operative treatment must be considered for each individual patient.
A number of factors are critical in deciding how to treat full-thickness tears, including a history of trauma, patient age, tear size, degenerative muscle and tendon changes, and functional disability.
Traditionally, open rotator cuff repair was the standard of care for symptomatic full-thickness rotator cuff tears.
Several disadvantages are inherent to open rotator cuff repairs. These include the need for deltoid detachment, difficult visualization of associated glenohumeral joint pathology, larger incisions, more extensive surgical dissection, and potentially a higher infection rate.
The surgical treatment of full-thickness rotator cuff tears has been revolutionized by the advent of arthroscopic surgery.
With the introduction of arthroscopy, rotator cuff repair has moved from mini-open repairs to complete arthroscopic repairs.
As techniques of complete arthroscopic rotator cuff repair have advanced, attempts have been made to treat larger tears arthroscopically. To do this, stronger fixation constructs must be used.
Single-row suture anchor repairs have been reported with good overall clinical results, but healing rates decrease as tear size increases.5,6
Double-row repair constructs with a medial and lateral row have been shown to provide improved initial biomechanical strength and restoration of the normal anatomic rotator cuff insertion.14–16
In the setting of a full-thickness rotator cuff tear, we now perform a double-row suture anchor repair if technically possible. While the double-row repair is more technically demanding, the potential advantages of anatomic restoration of the tendon insertion, improved biomechanical fixation, and improved healing may lead to improved functional outcomes.
ANATOMY
The rotator cuff is a complex of four muscles arising from the scapula and inserting onto the tuberosities of the proximal humerus.
The supraspinatus and infraspinatus muscles make up two thirds of the posterior cuff. The two tendons fuse together and have a direct bony insertion.
When performing a double-row rotator cuff repair, knowledge of the dimensions of the rotator cuff insertion or “footprint” is critical.
The supraspinatus averages 25 mm wide and has a medial-to-lateral footprint (tendon attachment) of 12.1 mm at the midtendon (FIG 1).
There is a normal sulcus between the articular cartilage and the medial aspect of the supraspinatus footprint; it averages 1.5 mm in width.
The infraspinatus has been shown to average 29 mm wide, with a mean medial-to-lateral width of 19 mm.
Suture anchor repair constructs using a single row of anchors have been shown to restore only 67% of the original footprint of the rotator cuff.2
Adding a second row of anchors increases the contact area of the repair 60%.21
The biomechanical properties of the double-row repair are improved compared to single-row repairs and include decreased strain over the footprint area, increased stiffness, and increased ultimate failure load.14
PATHOGENESIS
The etiology of rotator cuff tears is multifactorial.
The major factors are age-related degenerative changes of the tendon and physiologic loading.
The theory of age-related accumulative damage is supported by histologic findings of decreased fibrocartilage at the cuff insertion, decreased vascularity, fragmentation of the tendon with cellular loss, and disruption of Sharpey fiber attachments to bone.
Clinical studies support the aging theory as a primary cause of rotator cuff disorders.24
In a recent review of 586 consecutive patients with unilateral shoulder pain, rotator cuff tears were found to be correlated with increasing age, with an almost perfect 10-year difference between patients with no tear, a unilateral tear, and bilateral tears.
The average age of patients presenting with rotator cuffderived pain with no tear was 48.7 years old; unilateral tear, 58.7 years old; and bilateral tears, 67.8 years old.
Physiologic loading of the tendon has also been postulated as a mechanism for cuff tearing.
Localized degeneration of the articular region of the tendon, most commonly in the supraspinatus, is indicative of a tendon loading etiology.
FIG 1 • Anatomy of the rotator cuff footprint.
Uniform changes throughout the entire tendon, which are not commonly found, would be more suggestive of an agerelated degenerative process.
Age and loading likely have a multiplicative effect, with tendons in an older person both being more susceptible to damage from normal physiologic loading and exhibiting a worse healing response.
Genetics may also have a significant role in the predisposition for rotator cuff tears.
A strong relationship between rotator cuff tearing and family history has been shown.
One study found a relative risk of 2.42 for full-thickness rotator cuff tears in siblings of patients with cuff tears versus controls.10
This increased risk in siblings implies that genetic factors may play a role in the development of rotator cuff tears.
NATURAL HISTORY
Information about the natural history of rotator cuff disease is fundamental to understanding treatment indications.
Because symptomatic tears are often treated, our understanding of the natural history of rotator cuff disease is based on the study of asymptomatic rotator cuff tears.
Asymptomatic tears are extremely common in the population, and many of these are at risk for the development of symptoms over time.
In one study, over 51% of patients with a previously asymptomatic rotator cuff tear and a contralateral symptomatic tear will develop symptoms in the asymptomatic tear over an average of 2.8 years.25
Once a tear becomes symptomatic, 50% will progress in size. Only 20% of those remaining asymptomatic will progress.
No tears were found to decrease in size over time, suggesting that a significant percentage of patients with asymptomatic tears are at risk for symptom development.
Symptom development was also correlated with enlargement of the tear. Therefore, there is a limited intrinsic healing potential for the rotator cuff.
Just as importantly, there is a significant risk for tear progression, which will likely lead to significant functional deterioration and symptoms.
In addition, the potential for healing after surgery may be influenced by the irreversible muscle and tendon changes that occur in delayed repairs.
Clinical evidence of spontaneous healing of partial-thickness tears also appears limited.
Partial-thickness tears are likely to progress to full-thickness tears over time.
At an average of 8.4 years postoperatively, 35% of partial-thickness rotator cuff tears treated with arthroscopic acromioplasty and débridement without repair progressed to full-thickness tears.13
Although both partial-thickness and full-thickness tears tend to progress, the rate of progression in partial-thickness tears appears to be much slower.
PATIENT HISTORY AND PHYSICAL FINDINGS
Patients with rotator cuff disorders often complain of pain, weakness, or both in the shoulder.
The development of symptoms is often insidious.
There may be a recollection of minor trauma (eg, episode of heavy lifting, catching a heavy object).
Pain is usually localized to the anterior or anterolateral aspect of the shoulder, often extending down the front or side of the shoulder to the elbow.
Pain exacerbated with use, especially with overhead activities, is common.
Sleep disruption is also common in patients with symptomatic rotator cuff disease.
Weakness is a complaint for patients with large full-thickness rotator cuff tears.
Pain from tendinitis or small tears may simulate lack of strength, however. Therefore, weakness alone is not diagnostic of a large tear.
Similarly, patients with large or massive tears may have very reasonable function.
More commonly, however, these patients report overhead weakness and fatigue.
If gross weakness is recognized suddenly after a trauma, a rotator cuff injury should be suspected and investigated.
In the setting of chronic rotator cuff tears, inspection of the shoulder will often reveal atrophy of the supraspinatus and infraspinatus.
Prior surgical incisions should be noted. If previous open rotator cuff repair with deltoid detachment was performed, deltoid repair integrity should be assessed, along with axillary nerve function.
Range-of-motion testing should be performed both actively and passively.
Passive range of motion is often preserved except in the setting of chronic large tears, where static superior head migration leads to limited forward elevation with inferior capsule contracture.
Posterior capsular contracture is also a common finding with both small and large tears.
Active motion is often limited in scapular plane elevation. This may be due to either weakness or pain.
Shoulder strength should be evaluated with manual muscle testing.
Various arm positions will isolate the rotator cuff and specifically test these muscles for dysfunction.
The supraspinatus, infraspinatus, and teres minor can be isolated with resisted scapular plane elevation at 90 degrees in neutral rotation, resisted external rotation in full adduction and slight internal rotation, and external rotation in 90 degrees of abduction and 90 degrees of adduction, respectively.
Either the belly-press or lift-off test can be used to test subscapularis function.
Belly-press test: Inability to maintain maximum internal rotation without the elbow dropping posterior to the midsagittal plane of the trunk indicates impaired subscapularis function.
Lift-off test: Inability to maintain active maximal internal rotation with hand off the lumbar spine without extending the elbow indicates impaired subscapularis function.
Electromyographic analysis has shown that the belly-press activates the upper subscapularis while the lift-off activates the lower subscapularis.
Painful limitation of motion may limit the usefulness of the lift-off test.
This information may improve our ability to determine the extent of subscapularis dysfunction.
Special tests have been developed to aid in diagnosis:
The Neer impingement test (forward elevation in internal rotation) and the Hawkins impingement test (elevation to 90 degrees, cross-body adduction and internal rotation) were designed to elicit symptoms by impinging the rotator cuff on the undersurface of the acromion and coracoacromial ligament.
The Hornblower sign indicates teres minor dysfunction or tearing if there is weakness or inability to achieve full external rotation in an abducted position.
A positive result (weakness or pain) in the empty can test (Jobes sign) indicates dysfunction of the supraspinatus tendon.
Weakness with resisted external rotation in adduction represents dysfunction or tearing of the infraspinatus tendon.
External rotation lag sign: Inability to maintain the shoulder in a fully externally rotated position indicates significant dysfunction or tearing of the infraspinatus muscle.
Variable accuracy of these tests has been shown when used in isolation, but accuracy may be improved when used in combination with other provocative examinations.18
We do not routinely use these examinations, however. Instead, we base our findings on pain or weakness with resisted strength testing.
IMAGING AND OTHER DIAGNOSTIC STUDIES
Four standard shoulder radiographs should be taken of every patient evaluated for shoulder pain: anteroposterior (AP), true AP with active shoulder abduction, axillary lateral, and scapular-Y views.
The decision to obtain further imaging studies is based on radiographic findings along with data obtained from the history and physical examination.
In a patient with a small full-thickness rotator cuff tear, radiographs are usually normal.
With increasing tear chronicity, sclerotic and cystic changes of the greater tuberosity are often noted.
With increasing tear size, proximal humeral migration can be found on the AP and true AP views.
Proximal migration is best identified on the true AP view as loss of a concentric reduction of the proximal humeral and glenoid centers of rotation.
Humeral elevation may be static or dynamic depending on the chronicity of the tear. Static elevation occurs with contracture of the inferior capsule.
Narrowing of the acromiohumeral interval on the AP view has also been used to identify large tears.
MRI of the shoulder in patients with rotator cuff tears evaluates both the tendon and muscle quality.
Full-thickness tears show increased signal intensity at the tendon insertion on T2-weighted images.
MRI has been shown to have over 90% sensitivity and specificity in detecting tears without previous surgery.
Fatty infiltration and atrophy of the rotator cuff musculature can also be identified on MRI.
Increased fatty infiltration of the rotator cuff muscles has been correlated with poorer tendon healing and worse final postoperative outcomes after repair.
In the hands of a skilled ultrasonographer, ultrasound has a sensitivity and specificity similar to that of MRI.
Benefits of ultrasound include limited radiation, ability to routinely perform bilateral examinations, and a dynamic component of the examination, which can significantly aid in differentiating scar from tendon.
The most significant limitation of ultrasound is the need for an experienced ultrasonographer.
CT and CT arthrography has been widely used in Europe for the diagnosis of rotator cuff tears.
In patients with pacemakers or aneurysm clips, CT arthrography is a good alternative to MRI. Limitations of CT include increased radiation exposure and poorer soft tissue resolution compared to MRI.
Similar to MRI, muscle quality, including atrophy and fatty infiltration, can be examined and has been shown to be predictive of tendon healing and outcomes after surgery.
DIFFERENTIAL DIAGNOSIS
Rotator cuff tendinitis
Partial-thickness rotator cuff tear
Rotator cuff contusion
Adhesive capsulitis
Arthritis or chondral injury
Calcific tendinitis
Biceps tendon pathology (tendinitis or tearing)
Suprascapular nerve entrapment or spinoglenoid notch cyst
Internal impingement
NONOPERATIVE MANAGEMENT
The decision to pursue nonoperative management in the setting of a full-thickness rotator cuff tear depends on both patient and tear characteristics. Asymptomatic tears are extremely common, with MRI, ultrasound, and arthrography studies showing a 4% to 13% incidence in subjects 40 to 60 years old and over 50% in subjects older than 80.20 All asymptomatic tears should be treated nonoperatively. In subjects younger than 65, serial monitoring with sequential MRI or ultrasounds is reasonable, given that over 51% of patients with a previously asymptomatic tear and a contralateral symptomatic tear will develop symptoms in the asymptomatic shoulder over an average of 2.8 years.25
For symptomatic tears, nonoperative treatment has shown moderate success, with 45% to 82% satisfactory results.3,11,22 Nonoperative treatment includes anti-inflammatory medications, shoulder stretching, and rotator cuff and scapular stabilizer strengthening exercises. A limited number of subacromial cortisone injections may be performed, especially in patients who are not surgical candidates. Chronic large or massive rotator cuff tears in any age group or any chronic full-thickness tear in patients older than age 70 should undergo an initial trial (at least 3 months) of nonoperative management. Because irreversible changes have already occurred to the cuff or the articular cartilage in most of these patients, it is safe to attempt nonoperative treatment for a period of time. Failure of nonsurgical treatment is an indication for arthroscopic repair.
SURGICAL MANAGEMENT
The decision to proceed with operative treatment of rotator cuff disease requires an evaluation of the risks and benefits associated with both surgical and nonsurgical treatment.
While the risks of surgical management are well known, the risks of nonoperative treatment may not be so obvious.
Tear progression, muscle fatty infiltration and atrophy, and arthritis are potential irreversible risks of nonoperative treatment of rotator cuff tears.
Knowledge about these risks can help guide treatment.
Early surgical repair should be considered in all acute tears and any chronic small or medium-sized tears in patients younger than age 65.
These patients are at significant risk for developing the irreversible changes previously mentioned with prolonged nonoperative treatment.
These patients also have the greatest potential for healing. Consequently, the benefits of early surgical treatment combined with the inherent risks of prolonged nonoperative treatment guide us to early surgical repair.
Preoperative Planning
Tear size and chronicity will determine the difficulty of the repair, so careful preoperative imaging evaluation is important in surgical preparation.
If a tear is very large, the surgeon should make sure that a variety of different suture-passing devices are available to assist in the repair.
A Banana Suture Lasso (Arthrex, Naples, FL) can be passed through the Neviaser portal in large, medially retracted tears to shuttle suture through the tendon.
Angled Suture Lassos (Arthrex) can be placed through accessory portals to pass sutures from difficult angles not easily approached through the lateral working cannulas.
Larger anchors should be available if bone quality is poor.
Assess preoperative motion after the patient is anesthetized but before initiating the surgical procedure. Fixed superior humeral migration in the setting of a large rotator cuff tear will lead to inferior capsular contracture. In these patients, perform preoperative manipulation under anesthesia in forward elevation to increase the subacromial space available, thus facilitating the repair.
Positioning
Beach-chair position advantages
This is an anatomic position that permits better orientation and understanding of shoulder anatomy while performing the repair.
Examination under anesthesia is facilitated by stabilizing the scapula in the beach-chair position compared with the lateral position.
The arm can be easily manipulated in surgery without the need to unhook it from a traction unit.
Traction is not required but can be added in an inferior direction to increase the subacromial working space.
Humeral rotational control is easily accomplished. This can be critical when working on different regions (anterior vs. posterior) of the greater tuberosity.
Conversion to an open procedure is easily performed without redraping.
Lateral decubitus position advantages.
Many surgeons believe that the lateral position improves visualization and maneuverability of the scope due to traction.
It significantly improves inferior access to the glenohumeral joint, which makes it less difficult to perform glenohumeral procedures but has little impact on subacromial procedures.
Transient and permanent nerve damage has been reported due to traction in the lateral position. Consequently, we prefer to perform all subacromial procedures, including rotator cuff repair, in the beach-chair position.
TECHNIQUES
DOUBLE-ROW ROTATOR CUFF REPAIR WITH “MASON-ALLEN-TYPE CONSTRUCT USING SCREW-IN SUTURE ANCHORS
Portal Placement and Cannula Insertion
The camera is placed in the subacromial space through a posterior portal.
Our preferred starting posterior portal is slightly more lateral than a standard posterior portal. This is done to gain better visualization of the lateral greater tuberosity during repair. Also, a slightly inferior position is preferred, since portals will migrate superiorly with shoulder swelling.
A lateral working portal is developed under spinal needle localization. Portals should be placed low enough so that cannulas are introduced parallel to the rotator cuff tendon. This allows for easier subacromial instrumentation. The portal should be placed at about the midpoint of the tear in small or medium-sized tears.
A second lateral portal can be placed in larger tears with cannulas separated by several centimeters. Clear fully threaded 8.25–mm cannulas are placed in these portal sites.
Another large threaded cannula is placed through an anterolateral portal, anterior to the acromion, at the same level as the lateral and posterior portals. Again, maintaining low portal placement is critical so instruments will be passed parallel to the tendon, allowing the greatest excursion of instruments in the subacromial space. The anterolateral portal is mainly used as an accessory portal for suture retrieval and storage.
Repair Site Preparation
A soft tissue ablation device is used through the lateral portal to clear all the soft tissue on the undersurface of the acromion extending posteriorly, including the soft tissue and fat around the scapular spine. This will significantly improve the mobility of the tear. Soft tissue is removed from the greater tuberosity with a shaver, exposing cortical bone.
Mobility of the torn tendon is assessed with a tissue grasper through the lateral portal.
Anchor and Suture Placement
Once the tear has been determined to be repairable, a medial row of suture anchors (5.5-mm metal screw-in style) is placed. Anchors are loaded with two no. 2 Fiberwire sutures (Arthrex).
For small and medium-sized tears, we routinely place two medial anchors at the level of the anatomic neck. Each anchor is separated by 1 to 1.5 cm. Anchors are placed through small stab incisions just off the lateral border of the acromion.
For large and massive tears, we place three medial anchors.
Sutures from the medial row of anchors are next passed through the tendon. Starting with the most anterior anchor, both strands from one suture are passed through the tendon at the anterior aspect of the tear in a horizontal mattress fashion. Sutures are passed approximately 1 cm medial to the lateral edge of the tear. One strand of the second suture is passed adjacent to the most posterior strand of the first suture. This strand is retrieved out the anterolateral portal along with the two strands of the first suture.
The steps are repeated for the posterior anchor of the medial row. Two strands of one suture are passed at the posterior aspect of the tear in a mattress fashion. One strand of the second suture is placed just anterior to the previously placed mattress suture and retrieved out the anterolateral portal.
Both strands of the posterior mattress stitch are retrieved out the lateral portal, tied arthroscopically, and cut. Similarly, both strands of the anterior mattress stitch are retrieved out the lateral portal, tied, and cut.
The remaining strands passed through the tendon are in the anterolateral cannula and tied to one another outside the shoulder. The tails are cut and the knot is then advanced into the shoulder by pulling on the opposite two strands of the two sutures, creating a large horizontal mattress stitch between the anterior and posterior anchors (TECH FIG 1A).
A single lateral suture anchor is then placed at the lateral aspect of the rotator cuff footprint on the greater tuberosity, halfway between the medial anchors (TECH FIG 1B). One strand of one suture is retrieved out the lateral portal and passed medial to the horizontal stitch between the anterior and posterior medial anchors.
TECH FIG 1 • A. Arthroscopic picture showing the bridging horizontal mattress stitch between two medial-row anchors. Suture strands on right are the other limbs of the horizontal mattress stitch, which will be tied arthroscopically after the lateral-row stitches have been passed and tied. B. Lateral-row suture anchor placed at the lateral aspect of the greater tuberosity between the two medial-row anchors. C.Passage of lateral-row stitches medial to the bridging medial-row horizontal mattress stitch with a Scorpion suture passer. D. Final repair construct of a double-row repair using two medial and one lateral screw-in suture anchors.
This step is repeated with the second suture from the lateral anchor. These stitches are passed using a Scorpion suture passer (Arthrex) (TECH FIG 1C).
Once the lateral anchor sutures are passed, the remaining strands from the medial sutures are pulled on by an assistant to tension the medial horizontal mattress stitch between the medial anchors while the lateral-row sutures are tied. While tension is applied to the medial row, the lateral simple stitches are tied arthroscopically and cut.
Finally, the remaining two strands from the medial row anchors are retrieved out the lateral portal and tied arthroscopically.
The final construct has two medial-row anchors with a mattress stitch between the anchors and one lateral anchor with two simple stitches passed medial to the horizontal mattress between the medial anchors. This creates a “Mason-Allen” type of construct with the lateral simple stitches passed medial to the bridging medial horizontal mattress stitch (TECH FIG 1D).
DOUBLE-ROW ROTATOR CUFF TEAR WITH “MASON-ALLEN”-TYPE CONSTRUCT USING MEDIAL SCREW-IN SUTURE ANCHORS AND LATERAL PUSHLOCK ANCHORS
This repair technique follows the previous repair technique. After the anterior and posterior medial-row mattress sutures are tied, the tails of these sutures are not cut. Instead, they are retrieved out the anterolateral portal and stored. The bridging horizontal mattress stitch between the two medial anchors is created as described in the previous technique.
The untied suture strands (one from the anterior anchor and one from the posterior anchor) from the bridging mattress stitch between the anterior and posterior anchors (TECH FIG 2A) are retrieved out the lateral portal.
Both suture strands are then passed simultaneously medial to the horizontal mattress bridging mattress stitch with the Scorpion suture passer (Arthrex). This creates a Mason-Allen type of locking stitch construct.
Through one of the accessory lateral portals where the medial-row anchors were placed, a suture retriever is placed. Three strands are grabbed with the retriever: one strand from the tied posterior mattress stitch, one strand from the tied anterior mattress stitch, and one of the strands passed medial to the bridging horizontal mattress stitch.
All three strands are placed in a PEEK 3.5-mm PushLock knotless suture anchor (Arthrex). The PushLock awl is placed through the same lateral accessory portal and an anchor hole is tapped along the lateral aspect of the greater tuberosity at the posterior aspect of the tear (TECH FIG 2B). After a hole is tapped, the anchor is introduced into the joint through the same portal and impacted into the hole. The PushLock anchor has three strands (one strand from the tied posterior mattress stitch, one strand from the tied anterior mattress stitch, and one of the strands passed medial to the bridging horizontal mattress stitch) (TECH FIG 2C). As the anchor is impacted, all three strands should be tensioned to reduce the rotator cuff to the footprint. All three strands are then cut after final impaction of the PushLock anchor.
The previous steps are repeated, grabbing the second suture strand from the anterior and posterior mattress stitches and the second strand passed medial to the bridging mattress stitch. All three are placed in a second PushLock anchor and an anchor pilot hole is created at the anterior aspect of the tear along the lateral aspect of the greater tuberosity footprint.
TECH FIG 2 • A. Arthroscopic picture showing the opposite ends of the bridging horizontal mattress stitch shuttled through the cuff tendon medial to the bridging mattress stitch. B. A pilot hole is created using an awl for the posterior PushLock anchor as part of the lateral row. C. The posterior PushLock anchor is placed with one strand of suture from the anterior horizontal mattress stitch, one strand from the posterior horizontal mattress stitch, and one limb of the sutures from the bridging mattress stitch. D. The anterior PushLock anchor is placed, finishing the lateral row. E. Arthroscopic picture showing the final double-row construct using two lateral PushLock anchors and two medial screw-in anchors. F. Diagram showing the final double-row repair using two medial screw-in anchors and two lateral PushLock anchors.
After pilot hole creation, the anchor is introduced into the joint. With all three suture strands and while the strands are tensioned, the anchor is impacted (TECH FIG 2D). The tails of all three strands are then cut flush with the PushLock anchor, completing the repair.
The final construct consists of two medial-row anchors and two lateral-row PushLock anchors (TECH FIG 2E,F). Only two arthroscopic knots are required to complete this double-row repair.
POSTOPERATIVE CARE
All patients are initially placed in a sling, which is removed only for elbow range-of-motion exercises three or four times per day to limit elbow stiffness and for bathing.
We use a subacromial pain catheter, infiltrating 0.5% bupivacaine (Marcaine) during the first 48 hours postoperatively.
Patients remain on antibiotics (cephalexin) while the pain catheter is still in place.
Dressings are removed on the second postoperative day and showering is allowed the following day.
Patients are seen at 10 days postoperatively for suture removal.
When to start physical therapy after rotator cuff repair is debated among orthopedic surgeons. The decision is based on the perceived risks and benefits of early motion.
The major benefit of early motion is the potential limitation of postoperative shoulder stiffness. The main risks include repair disruption and limited healing.
Early passive motion has historically been recommended after open rotator cuff repair. With the advent of arthroscopic repairs, scarring from soft tissue dissection is minimized, so limiting early motion is possible.
Limited early motion may improve tendon healing.
Several factors, including tear size, tendon and bone quality, and preoperative motion, should be considered in this decision.
With osteoporotic bone or extremely poor tendon quality, limiting motion initially after repair is recommended.
Preoperative shoulder motion is an important factor in determining the initiation of motion. Earlier motion may be initiated if preoperative motion is limited and requires manipulation or release at the time of repair.
In general, tear size is the most important factor in determining the timing of postoperative rehabilitation.
Limiting early motion in patients with larger tears may provide improved healing potential, given that their overall healing rates are much lower than smaller tears.5–7
Patients with small or medium-sized tears remain in a sling for the first 6 weeks after surgery.
Elbow and hand range-of-motion exercises are started immediately.
No shoulder motion is allowed during the first 6 postoperative weeks.
If there was a significant preoperative motion deficit requiring surgical release or manipulation at the time of repair, early passive motion is allowed.
After 6 weeks, the sling is removed and patients are started on passive and active assisted range-of-motion exercises, including forward elevation in the scapular plane, external rotation in full adduction, and pendulum and pulley exercises.
Internal rotation and shoulder extension is limited and patients are instructed not to perform any lifting, pushing, pulling, or overhead activity.
At 3 months after surgery, strengthening exercises are initiated. These begin with isometric exercises and progress to isotonic exercises, with a stretching program maintained throughout.
Return to sports and full unrestricted activity is allowed at 4 to 5 months.
For large or massive tears, patients remain in a sling with no shoulder motion for 6 weeks.
At 6 weeks, the sling is removed and patients are allowed to lift the arm to shoulder height only.
Formal physical therapy is not initiated at this time. Instead, a shoulder continuous passive motion (CPM) device (Breg Flexmate S500, Breg, Inc., Vista, CA) is used to regain forward elevation in the scapular plane. CPM use is continued until 3 months postoperatively.
At this time, formal physical therapy is initiated, including passive and active motion and strengthening as per the protocol for small and medium-sized tears.
Return to sports and unrestricted activities is allowed at 6 months postoperatively.
OUTCOMES
Functional outcomes after both open and arthroscopic rotator cuff repair have been reported to be durable at long-term follow-up.7,23 A number of factors have been correlated with outcomes after repair, including patient age, tear size, tear acuity, workers' compensation status, preoperative smoking status, muscle quality, and tendon healing.
Most series reporting outcomes after complete arthroscopic rotator cuff repair are in single-row repairs. The potential advantage of a double-row repair is improved initial repair fixation strength and restoration of the normal anatomic rotator cuff footprint.14–16 Improved initial fixation strength and footprint restoration should lead to improved healing rates. In both open and arthroscopic repairs, tendon healing is correlated with improved outcomes.5,8,9Therefore, double-row repairs may lead to improved clinical outcomes.
There are limited series reporting the outcomes of complete arthroscopic double-row rotator cuff repairs.
Suguya et al19 compared healing rates and outcomes between singleand double-row repairs in 78 patients at an average of 35 months after surgery using MRI. There were significant improvements in UCLA and American Shoulder and Elbow Surgeons (ASES) scores in both repair groups, with no significant difference found between techniques. There was a significant increase in retear rates with single-row repairs.
Anderson et al1 recently evaluated 48 patients at a mean of 30 months after double-row repair with ultrasonography. There was a significant improvement in active motion, strength, and outcomes when compared to preoperative values. The overall retear rate was 17%, with no significant difference in outcomes between healed and retorn tendons. Healed shoulders were significantly stronger in elevation and external rotation.
Overall, double row-repairs appear to have improved healing rates compared to single-row repairs, although functional results are very similar.
COMPLICATIONS
Several factors can be directly correlated with persistent pain and limited function after repair.
These factors are broken down into three categories: surgeon-controlled, non-surgeon-controlled, and patientrelated factors.
They include incorrect or incomplete diagnosis, surgical technical error, stiffness, infection, and anesthesia-related complications.
Continued pain after rotator cuff repair can often occur if a second pathology is not identified and treated.
Conditions often confused with rotator cuff disease include cervical spine disorders, suprascapular neuropathy, acromioclavicular joint arthritis, biceps tendonopathy, glenohumeral instability or arthritis, labral tears, and frozen shoulder.
A complete history and physical examination can prevent missing several of these problems, which can often be treated concomitantly at the time of rotator cuff repair.
Technical problems leading to persistent pain and dysfunction after repair can be grouped into repair failures, deltoid detachment, neurologic injury, excess fluid extravasation, and patient positioning injuries.
The most likely reason for failure of tendon healing after repair is patient age.
Poor surgical technique, including poor knot-tying, limited fixation (number of anchors), and poor anchor insertion technique, can all lead to a weak biomechanical construct.
Deltoid detachment is avoided in the setting of complete arthroscopic repair, but if a mini-open approach is performed, then excess detachment without bony repair can lead to failure of healing.
Transient neurologic injury can occur secondary to excess traction when the lateral position is used.
Proper portal placement is critical to avoid axillary (posterior and lateral portals) and musculocutaneous (anterior portal) nerve injury.
Excess swelling due to fluid extravasation into the deltoid can significantly raise intramuscular pressures. Therefore, pump pressures should be kept below 50 mm Hg, with procedure times less than 2 hours.
Proper padding around the knees (lateral position) and flexing the hips and knees (beach-chair position) can avoid iatrogenic problems secondary to positioning.
Postoperative stiffness is another potential complication.
With limited surgical dissection associated with complete arthroscopic repairs, the risk of stiffness may be significantly reduced when compared with open repairs.
While overall rates of postoperative stiffness have not been clearly reported, more than 5% to 10% of open repairs are complicated by either adhesions in the humeral scapular interface or capsular contracture.
We now routinely hold all shoulder motion after arthroscopic repairs for several weeks in an attempt to improve healing rates, with limited concern for developing postoperative stiffness.
If significant stiffness does develop that is resistant to therapy, arthroscopic lysis of adhesions in the subacromial space along with capsular release is recommended.
Infection after rotator cuff repair is uncommon.
Most series report infection rates of 1% to 2% after open or mini-open rotator cuff repairs.
While there are few reported studies of infection rates after complete arthroscopic repairs, it appears that infection is less common than after open or mini-open repairs.
Diagnosis is often delayed in cases of postoperative infection, and persistent wound drainage is the most consistent examination finding.
Cultures will often grow Propionibacterium acnes, Staphylococcus aureus, and coagulase-negative Staphylococcus aureus.
P. acnes often takes 7 to 10 days to grow on cultures. Therefore, cultures should be held in the setting of postoperative infections for at least 1 week.
Treatment consists of multiple débridements and intravenous antibiotics for usually 6 weeks.
Outcomes after infection are satisfactory, although significant delays in diagnosis or treatment can lead to inferior results.
Anesthetic complications can occur after rotator cuff repair.
If general anesthesia is used, major complications occur less than 1% of the time.4
More commonly, nausea, inability to void, and severe pain are the complications seen in the setting of outpatient elective shoulder surgery.
If an interscalene block is used, inadequate anesthesia is the most common complication.
Temporary Horner syndrome, phrenic nerve paralysis, and recurrent laryngeal nerve block are common but usually without significant consequence.
Intraneural injection or needle injury to the nerve roots can occur.
Symptoms such as persistent paresthesias or numbness can be irritating but usually resolve with time (possibly several months).
REFERENCES
1. Anderson K, Boothby M, Aschenbrener D, et al. Outcome and structural integrity after arthroscopic rotator cuff repair using 2 rows of fixation: minimum 2-year follow-up. Am J Sports Med 2006;34:1899–1905.
2. Apreleva M, Ozbaydar M, Fitzgibbons PJ, et al. Rotator cuff tears: the effect of the reconstruction method on the three-dimensional repair site area. Arthroscopy 2002;18:519–526.
3. Bartolozzi A, Andreychik D, Ahmad S. Determinants of outcome in the treatment of rotator cuff disease. Clin Orthop Relat Res 1994; 308:90–97.
4. Bishop J, Klepps S, Lo IK, et al. Cuff integrity after arthroscopic versus open repair: a prospective study. J Shoulder Elbow Surg 2006;15: 290–299.
5. Boileau P, Brassart N, Watkinson DJ, et al. Arthroscopic repairs of full-thickness tears of the supraspinatus: does the tendon really heal? J Bone Joint Surg Am 2005;87A:1229–1240.
6. Galatz LM, Ball CM, Teefey SA, et al. The outcomes and repair integrity of completely arthroscopically repaired large and massive rotator cuff tears. J Bone Joint Surg Am 2004;86A:219–224.
7. Galatz LM, Griggs S, Cameron BD, et al. Prospective longitudinal analysis of postoperative shoulder function: a ten-year follow-up study of full-thickness rotator cuff tears. J Bone Joint Surg Am 2001; 83A:1052–1056.
8. Gazielly DF, Gleyze P, Montagnon C. Functional and anatomic results after rotator cuff repair. Clin Orthop Relat Res 1994; 304:43–53.
9. Harryman DT, Mack LA, Wang KY, et al. Repairs of the rotator cuff. Correlation of functional results with integrity of the rotator cuff. J Bone Joint Surg Am 1991;73A:982–989.
10. Harvie P, Ostlere SJ, The J, et al. Genetic influences in the aetiology of tears of the rotator cuff: sibling risk of a full-thickness tear. J Bone Joint Surg Br 2004;86B:696–700.
11. Itoi E, Tabata S. Conservative treatment of rotator cuff tears. Clin Orthop Relat Res 1992;275:165–173.
12. Jobe FW, Moynes DR. Delineation of diagnostic criteria and a rehabilitation program for rotator cuff injuries. Am J Sports Med 1982; 10:336–339.
13. Kartus J, Kartus C, Rostgard-Christensen L, et al. Long-term clinical and ultrasound evaluation after arthroscopic acromioplast1y in patients with partial rotator cuff tears. Arthroscopy 2006;22: 44–49.
14. Kim DH, Ellatrache NS, Tibone JE, et al. Biomechanical comparison of a single-row versus double-row suture anchor technique for rotator cuff repair. Am J Sports Med 2006;34:407–414.
15. Ma CB, Comerford L, Wilson J, et al. Biomechanical evaluation of arthroscopic rotator cuff repairs: double-row compared with singlerow fixation. J Bone Joint Surg Am 2006;88A:403–410.
16. Mazzocca AD, Millett PJ, Guanche CA, et al. Arthroscopic singlerow versus double-row suture anchor rotator cuff repair. Am J Sports Med 2005;33:1861–1868.
17. Miniaci A, Dowdy PA, Willits KR, et al. Magnetic resonance imaging evaluation of the rotator cuff tendons in the asymptomatic shoulder. Am J Sports Med 1995;23:142–145.
18. Park HB, Yokota A, Gill HS, et al. Diagnostic accuracy of clinical tests for the different degrees of subacromial impingement. J Bone Joint Surg Am 2005;87A:1446–1455.
19. Suguya H, Maeda K, Matsuki K, et al. Functional and structural outcome after arthroscopic full-thickness rotator cuff repair: single-row versus dual-row fixation. Arthroscopy 2005;21:1307–1316.
20. Tempelhof S, Rupp S, Seil R. Age-related prevalence of rotator cuff tears in asymptomatic shoulders. J Shoulder Elbow Surg 1999;8: 296–299.
21. Tuoheti Y, Itoi E, Yamamoto N, et al. Contact area, contact pressure, and pressure patterns of the tendon–bone interface after rotator cuff repair. Am J Sports Med 2005;33:1869–1874.
22. Wirth MA, Basamania C, Rockwood CA. Nonoperative management of full-thickness tears of the rotator cuff. Orthop Clin North Am 1997;28:59–67.
23. Wolf EM, Pennington WT, Agrawal V. Arthroscopic side-to-side rotator cuff repair. Arthroscopy 2005;21:881–887.
24. Yamaguchi K, Ditsios K, Middleton WD, et al. The demographic and morphologic features of rotator cuff disease: a comparison of symptomatic and asymptomatic shoulders. J Bone Joint Surg Am 2006;88A:1699–1704.
25. Yamaguchi K, Tetro AM, Blam O, et al. Natural history of asymptomatic rotator cuff tears: a longitudinal analysis of asymptomatic tears detected sonographically. J Shoulder Elbow Surg 2001;10:199–203.