Management of rotator cuff tears

INTRODUCTION — The rotator cuff is a group of four muscles and their associated tendons that originate from locations on the scapula and insert onto the humeral head. These four muscles (supraspinatus, infraspinatus, subscapularis, and teres minor) stabilize the glenohumeral joint, enable rotation around the joint, and provide a counterforce to the larger, more superficial muscles that exert forces at the shoulder (eg, deltoid) [1-3].

The management of rotator cuff tears is reviewed here. The presentation and diagnosis of rotator cuff tears, diagnosis and management of rotator cuff tendinopathy and shoulder impingement, and diagnostic approach to shoulder pain more generally are all discussed separately. (See "Presentation and diagnosis of rotator cuff tears" and "Rotator cuff tendinopathy" and "Subacromial (shoulder) impingement syndrome" and "Evaluation of the adult with shoulder complaints".)

DEFINITIONS AND TEAR TYPES — Rotator cuff tears may be acute, subacute, or chronic [4]. They involve a tear of the tendon, musculotendinous junction, or muscle or an avulsion from bone. Tears that develop parallel to tendon and muscle fibers (longitudinal tears) produce less weakness and disability. Transverse tears across tendon or muscle fibers cause greater weakness and may lead to retraction of muscle tendon fibers. Increased fatty infiltration and muscle atrophy can be seen in chronic tears.

While there is no standard classification scheme, most tears are categorized as full thickness (extending completely from the articular to the bursal surface) or partial thickness. Partial-thickness tears are described as either articular sided or bursal sided, with articular-sided tears being two to three times more common.

Tear size is described as small, medium, large, or massive. Massive tears often involve multiple tendons.

●Small: <1 cm

●Medium: 1 to 3 cm

●Large: 3 to 5 cm

●Massive: >5 cm [5]

Tear characteristics associated with impaired healing include:

●Longer duration (ie, older tears)

●Transverse (versus longitudinal)

●Greater retraction

●More extensive muscle atrophy and fatty infiltration (especially for chronic degenerative tears) [6]

Comorbidities associated with an increased risk for rotator cuff disease include elevated cholesterol, diabetes mellitus, and obesity. These risk factors are discussed in greater detail separately. (See "Presentation and diagnosis of rotator cuff tears", section on 'Risk factors'.)

MANAGEMENT AND EVIDENCE LIMITATIONS — The treatment of rotator cuff tears depends upon several factors, including the duration of symptoms (acute versus chronic), type of tear (eg, partial thickness, full thickness, degenerative), clinical findings (ie, pain and degree of disability), and patient characteristics (eg, age, comorbidities, and activity level) [7-12]. Treatment consists primarily of physical therapy or surgical repair. Our approach to management is summarized in the following flow chart (algorithm 1).

To date, few randomized trials have been performed that directly compare surgical and nonsurgical management of rotator cuff tears [7,8]. Of the studies performed, many have not demonstrated a clear benefit to surgery [13-16]. Two systematic reviews of studies evaluating common interventions for the treatment of rotator cuff tears, including studies comparing different surgical approaches and others comparing surgery with physical therapy, found there was insufficient evidence to support one intervention over another and found the quality of many studies to be poor [8,17]. A subsequent systematic review limited to outcomes in patients (n = 4542) with full-thickness tears participating in randomized trials noted consistent improvements in shoulder symptoms and function at one-year follow-up regardless of treatment approach [16].

One important limitation of many studies of rotator cuff tear management is the duration of follow-up. The benefit of early surgical repair may not be appreciated in studies where follow-up is limited to one or two years. As an example, in a well-conducted randomized trial comparing primary tendon repair and physical therapy for small and medium-sized (<2.5 cm), full-thickness rotator cuff tears, patients managed surgically had substantially better outcomes at ten-year follow-up as determined by validated function scores (Constant, American Shoulder and Elbow Surgeons), pain assessment, and mobility testing than patients managed with physical therapy [18]. The patients managed with physical therapy experienced a gradual functional decline after the two-year follow-up, which may have been due to increasing tear size and fat atrophy of the muscle-tendon unit [19].

Given the dearth of high-quality evidence and the equivocal results of many studies, the approach to the management of rotator cuff tears discussed below is based largely upon clinical experience, observational data, animal studies, and extrapolations from studies of the treatment of other tendon tears [9,20-24].

MANAGEMENT BY CLINICAL PRESENTATION AND TEAR TYPE

Acute traumatic tear

Initial interventions, younger patients, terminology — Acute rotator cuff tears are sustained during sport, often in overhead athletes, or from trauma and present with a sudden onset of pain and possibly weakness in a previously asymptomatic shoulder [25]. Acute tears may be partial thickness, extending only partway through the tendon, or full thickness [26]. Magnetic resonance imaging (MRI) should be obtained within approximately two weeks of injury for patients suspected of having sustained an acute, traumatic rotator cuff tear. If MRI reveals signs of such an injury, the patient should be referred to a shoulder surgeon. Patients without signs of tear but whose clinical presentation suggests rotator cuff injury (eg, tendinopathy) are referred to physical therapy. (See "Presentation and diagnosis of rotator cuff tears" and "Rehabilitation principles and practice for shoulder impingement and related problems".)

Initial care includes rest, ice, and oral analgesics for pain control. Ice may be applied for 15 minutes at a time every one to two hours for the first day or two after injury. A cloth or towel should be used to prevent direct contact between ice and skin. Acetaminophen and nonsteroidal antiinflammatories are both acceptable options for analgesia. Overhead movements and heavy lifting should be avoided to prevent worsening pain and further injury. Keeping the elbows close to the side while lifting, pushing, or pulling allows patients to tolerate most movements without difficulty. This includes basic activities such as transferring from a sitting to standing position.

Younger patients (<40 years old) with acute traumatic tears typically do better with early arthroscopic or open repair performed within about eight weeks of the injury. These injuries tend to occur in younger patients and involve less muscle atrophy, fatty infiltration, and tendon retraction than chronic, nontraumatic tears. Thus, they have better outcomes with surgical repair than chronic tendon injuries [25,27]. In a systematic review of 12 studies involving patients between 16 and 40 years of age with rotator cuff tears (n = 336), surgical repair led to substantial improvements in pain and function in the large majority [27]. However, among a subgroup of elite throwers, many were unable to achieve their prior level of performance. Therefore, many overhead athletes are best treated with physical therapy, possibly in combination with limited surgical debridement.

A systematic review of 46 studies found no consensus on the definition of "acute" or "traumatic" rotator cuff tears. The authors' recommendation was that the term “acute” be reserved for tears showing edema, wavelike appearance of the central portion of the torn tendon, and joint effusion on MRI. To reveal these findings, MRI should be obtained within about two weeks of the injury. The authors recommended that the term “traumatic” be used for patients who experience a sudden onset of symptoms (eg, pain, weakness) in a previously asymptomatic shoulder associated with a consistent traumatic event (eg, fall onto retroverted arm causing shoulder dislocation) [25].

Partial thickness — Regardless of patient age, most partial-thickness tears merit a trial of physical therapy before surgery is considered. This holds true regardless of whether tears are transverse or longitudinal. Exceptions are individuals with severe weakness or functional deficits and those needing to return to a physically demanding sport, activity, or work. These patients should be referred to a shoulder surgeon, as repair or debridement may be needed. Patients with severe pain but only mild functional deficits may benefit from a trial of physical therapy once their pain is adequately controlled with oral medication or early glucocorticoid injection. (See 'Physical therapy' below and 'Indications for surgical referral' below.)

There appears to be no harm and possibly benefit from managing patients initially with physical therapy and careful monitoring and then proceeding to surgical repair as needed. This approach is supported by the results of a randomized trial of 78 consecutive patients with partial-thickness tears of the supraspinatus in which patients treated with delayed repair had better functional outcomes at six months compared with those treated with immediate surgical repair [28].

For the first six weeks following injury, patients with acute partial-thickness tears should use caution when lifting or performing extensive work involving the injured arm other than rehabilitation exercises. Daily activities that do not cause discomfort need not be restricted. After six weeks, activity involving the arm may be increased gradually under the guidance of the physical therapist.

Partial-thickness tears may enlarge over weeks to years, and some may develop into full-thickness tears. Thus, such tears should be monitored during physical therapy and thereafter for signs of enlargement, which may include increasing weakness and worsening pain. (See 'Monitoring partial-thickness and small full-thickness tears' below.)

Full thickness — Small (<1 cm), full-thickness tears have a slower rate of progression and may remain stable, more like partial-thickness tears, and many can be treated with physical therapy. Patients whose function and symptoms do not improve with three months or more of appropriate physical therapy or worsen after the initial six weeks of physical therapy are referred to a shoulder surgeon. (See 'Physical therapy' below and 'Indications for surgical referral' below.)

Larger (>1 to 1.5 cm) full-thickness tears have a high rate of tear progression [29,30]. Patients with such injuries should be referred to a shoulder surgeon. Full-thickness tears with significant retraction and those that are large or massive in individuals who are otherwise good surgical candidates are generally treated surgically.

Patients will full-thickness tears who are treated initially with physical therapy should avoid activities requiring heavy lifting or extensive work with the injured arm outside of their rehabilitation exercises. Other daily activities, such as computer work, are typically not restricted.

Overhead athletes — Overhead athletes with rotator cuff tears constitute a unique subgroup. Their injuries often result from repetitive microtrauma and shoulder impingement leading to partial-thickness tears that can progress to full thickness. Even for many younger overhead athletes, physical therapy is a reasonable initial approach to treatment, as surgical outcomes are not ideal. When pain rather than tear size is the major problem, less invasive surgery such as debridement may be helpful if physical therapy (possibly with glucocorticoid injection) is not effective. (See 'Overhead athletes' below and 'Glucocorticoid injection' below.)

Acute-on-chronic tear — In an acute-on-chronic tear, retraction of the tendon by 2 cm or more or weakness that is new or worsening are reasonable indications for referral to orthopedic surgery. Nevertheless, patients with significant comorbidities may not be suitable surgical candidates. (See 'Effect of tear characteristics and comorbidities on outcomes' below.)

Patients with these injuries should refrain from lifting or performing repetitive work involving the injured arm for several weeks until pain subsides. Daily activities that can be done without pain may be performed.

Massive tear — Massive rotator cuff tears involve extensive tendon injury and can be complicated by functional failure of the rotator cuff. All patients with such tears should be referred to a shoulder surgeon. If the patient is a reasonable surgical candidate and the tear appears reparable on advanced imaging, most shoulder surgeons advocate operative repair for these injuries.

The criteria defining a massive tear remain an area of debate, and several definitions have been proposed. These include tears of two or more tendons, tear size greater than 5 cm in an anterior-posterior or medial-lateral orientation, or retraction of the tendon to the glenoid rim with greater than two-thirds of the greater tuberosity exposed on imaging in the sagittal plane [31,32].

Massive tears are classified as reparable or irreparable, although the criteria for this distinction are not firmly established. Suggested criteria for irreparable tears include [33-37]:

●Substantial fatty degeneration of muscle (Goutallier stage 3 or 4; ie, 50 percent or greater)

●Tendon retraction (Patte stage 3, ie, proximal tendon stump at level of glenoid or greater retraction)

●Muscle atrophy (grade 3 [severe] based on MRI appearance)

●Acromiohumeral distance of less than 6 to 7 mm

Chronic degenerative tear — As people age, chronic degeneration of tissue makes tendons more susceptible to tearing [25]. Advanced age, multiple comorbidities (eg, diabetes), and extensive fatty infiltration of the rotator cuff increase this susceptibility. Regardless of baseline function or comorbidities, initial treatment of degenerative tears consists of pain relief, physical therapy, and activity modification. As the available evidence suggests that conservative and surgical management result in similar clinical outcomes for these injuries, every effort should be made to maximize the results of physical therapy before surgery is considered in older adults with degenerative tears [38].

SURGICAL MANAGEMENT

Indications for surgical referral — Most rotator cuff tendon tears are treated initially with physical therapy, and surgery is considered only when conservative measures fail. However, some injuries warrant immediate surgical referral, including the following:

●Acute, full-thickness, traumatic tear, especially in younger (<40 years), otherwise healthy patients – Such injuries are usually treated with immediate surgery, as delay can lead to significant muscle atrophy, tendon retraction, and poorer surgical results.

●Sudden or severe weakness (eg, inability to abduct or flex shoulder) in patient with pre-existing, partial-thickness tear – These findings are consistent with the development of a full-thickness tear, and urgent surgical repair may be needed depending on such factors as occupation and patient age.

●Failure to improve after three months of well-designed and properly performed physical therapy, or symptoms or function worsen during the initial six weeks of physical therapy.

In addition to the major categories above, we commonly make an early referral for specific patient populations so they can discuss the relative risks and benefits of repair with a shoulder surgeon, even if surgery is not the best option. These populations include:

●Heavy laborer with partial tear who depends on a functional shoulder for their livelihood

●Elite or professional overhead athlete

Following acute, traumatic tears that produce significant functional deficits, MRI should be obtained to determine the extent of injury and need for surgery. In experienced hands, diagnostic ultrasound provides similar information, and a dynamic examination can show the degree of tear retraction. Imaging is discussed separately. (See "Presentation and diagnosis of rotator cuff tears", section on 'Imaging studies'.)

Increased retraction of a torn tendon or tendons is an indication for surgery. Greater than 80 percent of symptomatic, full-thickness tears and 25 percent of partial-thickness tears increase in size [39]. The rate of enlargement is faster in full-thickness tears. If a tear amenable to surgery is identified, outcomes are better when the repair is performed within eight weeks of injury [25]. Partial-thickness tears should be monitored for propagation of the tear, which may necessitate surgical referral depending on the extent. (See 'Monitoring partial-thickness and small full-thickness tears' below.)

Surgical procedures and postoperative timeline — When indicated, surgical options include arthroscopic repair or debridement, open rotator cuff repair, and shoulder arthroplasty [40]. Selection depends upon the nature of the injury. There is significant variation among orthopedic surgeons in the management of rotator cuff tears [41]. Common approaches include the following:

●Smaller partial-thickness tears may do well with debridement when surgery is indicated. While debridement of damaged rotator cuff tissue, with or without subacromial decompression, remains a common approach for smaller, partial-thickness tears, several studies have failed to demonstrate a clear benefit from the addition of subacromial decompression [8,42-44].

●Larger partial-thickness tears can be repaired using a trans-tendon technique or by converting the tear to a full-thickness tear and then repairing it [29].

●Massive rotator cuff tears may be treated with a superior capsule reconstruction, tendon transfer in younger, active patients, or reverse total shoulder arthroplasty in older, relatively sedentary adults (particularly those with osteoarthritis of the glenohumeral joint) [31].

●Reverse total shoulder arthroplasty is the most common surgical intervention for rotator cuff arthropathy in the setting of glenohumeral osteoarthritis. However, there are little data comparing this technique to other forms of shoulder replacement [45].

The timeline for immobilization and rehabilitation following surgery varies by injury, procedure, and individual surgeon protocols. Evidence is limited, and the following are rough estimates. Following debridement, the shoulder is usually immobilized in a sling for two to three weeks, but early passive range of motion may be done and sling use may be continued for comfort if needed beyond three weeks. Full mobility is usually achieved by six weeks. Following surgical repair of the rotator cuff, the shoulder is commonly immobilized (often in an abduction brace) for six weeks, and maximal mobility is achieved by four to five months, while regaining maximal function may require six months or longer. Arthroplasty follows a similar timeline as repair, but some preliminary studies suggest that outcomes may be improved with gentle, early mobility exercise.

Surgical outcomes

Massive tear — Massive rotator cuff tears appear to have better functional outcomes with surgical repair compared with physical therapy alone. However, evidence supporting this approach is limited largely to observational studies and clinical experience [8,46].

Surgical repair of massive tears usually results in improvement in overall mobility and function. A systematic review of nine studies comparing different surgical treatments reported improvements in active range of motion, with over 80 percent of patients achieving greater than 90 degrees of forward flexion regardless of surgical technique [47]. However, the overall increase in mobility varied substantially by study and procedure.

Full-thickness tear — The results of studies assessing surgical outcomes for full-thickness tears are mixed. A meta-analysis of six randomized trials assessing the management of full-thickness tears reported no significant differences in functional outcome at one- and two-year follow-up between those treated surgically and those treated with physical therapy [48]. Several studies did not report outcome by tear type. A retrospective cohort study of 559 full-thickness rotator cuff tears reported that patients treated surgically did better across all outcome measures [49]. However, these findings may be limited by the relatively short follow-up period, among other problems.

In a well-conducted randomized trial comparing primary tendon repair and physical therapy for small and medium-sized (<2.5 cm), full-thickness rotator cuff tears, patients managed surgically had substantially better outcomes at ten-year follow-up as determined by validated function scores (Constant, American Shoulder and Elbow Surgeons), pain assessment, and mobility testing than patients managed with physical therapy [13,18,50]. The patients managed with physical therapy experienced a gradual functional decline after the two-year follow-up assessment, which may have been due to increasing tear size and fat atrophy of the muscle-tendon unit [19]. While further studies are needed to corroborate these findings, the results underscore the importance of close surveillance and early surgical referral, generally within six months of presentation, for young or active patients with full-thickness rotator cuff tears and persistent symptoms.

Partial-thickness tear — Studies included in a systematic review of surgical repair for partial-thickness rotator cuff tears (n = 657 shoulders) reported a wide range of outcomes, with "excellent" results found in 29 to 93 percent of cases depending upon tear size and operative technique [42]. Procedures included debridement (with or without subacromial decompression), trans-tendon repair, and takedown with repair. Tears extending less than 50 percent of tendon depth had favorable outcomes with debridement, with or without subacromial decompression. Tears extending beyond 50 percent of tendon depth did better with repair, although evidence was insufficient to support any particular technique.

Overhead athletes — Overhead athletes with rotator cuff tears are usually treated with physical therapy, and surgery is reserved for tears that fail to improve. In such cases, debridement is the preferred approach, as many high-level athletes are unable to return to their previous level of performance following full surgical repair despite immediate symptomatic improvements postoperatively [27,51-53]. (See 'Physical therapy' below.)

In a 2016 meta-analysis including 635 athletes (286 professional or elite) from 23 studies, all treated surgically for rotator cuff tear, return to sport was achieved by 84.7 percent overall (95% CI 77.6-89.8), but only 49.9 percent (95% CI 35.3-64.6%) returned to their preinjury level of play [51]. Among the overhead athletes included in a retrospective study of 32 consecutive adolescents treated with surgical repair of rotator cuff tears, 13 of 14 were able to return to the same level of play, but eight could not continue to play their preinjury position (eg, pitcher) [52].

Effect of tear characteristics and comorbidities on outcomes — Certain tear characteristics and patient comorbidities are associated with significantly higher rates of surgical complications when undergoing rotator cuff repair surgery. Such patients are often better suited to conservative management, at least initially.

Factors associated with an increased risk of failure of the surgical repair include the following [49,54-57]:

●Increased fatty infiltration of the tendon

●Large or massive tear in patient with glenohumeral osteoarthritis

●History of smoking

●Diabetes mellitus

NONOPERATIVE MANAGEMENT

Physical therapy — Well-designed and appropriately implemented physical therapy helps most patients with degenerative or acute, partial-thickness tears to improve their shoulder function and reduces symptoms. Evidence suggests that any tear progression that might occur is minimal over the timeframe needed to determine whether physical therapy will be effective.

While a variety of rehabilitation protocols are used for rotator cuff tears, the underlying principles among them remain consistent. Overall, the main focus is on scapular stabilization and appropriate, progressive loading of the rotator cuff tendons based on the phase of healing [58]. The principles and implementation of physical therapy for rotator cuff disorders are reviewed in detail separately. (See "Rehabilitation principles and practice for shoulder impingement and related problems".)

The phases of rotator cuff healing and the associated stages of rehabilitation can be summarized as follows [59]:

●Acute phase – The acute phase is focused on reducing pain and inflammation while protecting the tendon from further damage. As pain improves, gentle stretching and strengthening exercises to prevent atrophy and contractures are introduced. These include passive range of motion in patients with pain and active range of motion when pain improves. Scapulothoracic mobility is also addressed.

●Recovery phase – The recovery phase focuses on strengthening the scapular stabilizers and rotator cuff as well as improving any kinetic chain deficits. Without scapular stabilization, strengthening of the rotator cuff may fail.

●Functional phase – The functional phase includes more advanced strength exercises including eccentric exercises. This phase requires painless, near-full range of motion.

●Sport-specific phase – The final phase consists of sport- or activity-specific training that enables the athlete or heavy laborer to withstand the stresses and loads they will endure upon their return.

In addition to progressing through the stages outlined above, we believe that patients who have successfully completed a course of physical therapy and have resumed their usual activities and patients with tears that appear to be stable should continue to perform functional strength and mobility exercises. Although formal evidence supporting this approach is limited, we believe that continued performance of strength exercises for the scapular stabilizers and rotator cuff and mobility exercises to maintain proper motion of the scapula and glenohumeral joint help to reduce the risk for recurrence or worsening injury.

Properly performed physical therapy does not appear to increase the rate of tear propagation or otherwise exacerbate injury, and thus, most rotator cuff tears are amenable to a trial of physical therapy prior to surgical intervention. In a systematic review of eight studies involving 411 asymptomatic and symptomatic full-thickness rotator cuff tears, there was no difference in the rate of tear progression over time, with little significant progression over the short to intermediate timeframe [60]. Comparison of physical therapy with surgical repair in a small randomized trial of adults with acute supraspinatus tears found no significant difference in tear progression [61]. In a prospective cohort study of 452 patients with a full-thickness rotator cuff tear, over 75 percent experienced symptomatic improvement that persisted beyond two years [62]. Patients that did not improve and opted for surgery made their decision within the first 6 to 12 weeks [62].

Future directions in therapy for rotator cuff tears include the potential for use of wearable technology and inertial measurement units to monitor both movement and posture during rehabilitation [58].

Glucocorticoid injection — Glucocorticoid injection is commonly used to treat the symptoms (primarily pain) caused by a rotator cuff tear. In particular, we believe it is reasonable to treat patients whose pain prevents them from participating in physical therapy with such an injection. We do not give a glucocorticoid injection to patients if a significant rotator cuff tear that may require surgery is known or suspected. A single injection at a minimum of three months prior to arthroscopic repair is reasonable for patients requiring pain relief. Extending that interval to six months may further decrease the risk for operative complications.

While the goal of these injections is to reduce inflammation and pain and thereby improve function, the benefits are uncertain [63]. The results of a few small, randomized trials suggest some short-term benefit [64,65]. However, when added to a progressive physical therapy program or a single session of physical therapy plus best practice advice, glucocorticoid injection did not improve outcomes at 12 months in a randomized trial of 708 patients with rotator cuff disorders [66]. These results are consistent with other studies of glucocorticoid injection for muscle and tendon injury. (See "Overview of the management of overuse (persistent) tendinopathy", section on 'Glucocorticoids'.)

Glucocorticoid injections entail some risk. Tendon and fascial tissue injuries have been reported following injections for athletic injuries [67]. Numerous studies report an increased risk of rotator cuff repair failure necessitating revision surgery following glucocorticoid injection in the perioperative setting [68-71]. Multiple injections or injections closer to the time of arthroscopic surgery appear to increase this risk.

Unproven and investigational adjunct therapies

●Platelet-rich plasma therapy — Studies of platelet-rich plasma (PRP) for the treatment of rotator cuff tear report mixed results. Overall, there is insufficient high-quality evidence to support PRP as a mainstay of therapy for rotator cuff tears, and PRP is not part of our standard approach to treatment. Clinicians who use PRP are obligated to share with patients the limited evidence for clinically meaningful benefit, associated risks, and costs of treatment. PRP in the treatment of muscle and tendon injury is discussed in greater detail separately. (See "Biologic therapies for tendon and muscle injury".)

A 2020 meta-analysis of 17 randomized trials (1116 patients) found that PRP therapy led to reduced long-term retear rates and statistically significant improvements in multiple pain and function scores, but none achieved minimal clinically important differences [72]. The authors concluded that "PRP may positively affect clinical outcomes, but limited data, study heterogeneity, and poor methodological quality hinder firm conclusions." Studies of PRP as adjunctive therapy following rotator cuff repair have shown a lack of efficacy [73,74].

Long-term improvement in pain and function with short- and medium-term improvement in function has been shown with PRP injections in one systematic review comparing various injections for rotator cuff tears [75].

●Prolotherapy — A 2020 systematic review found too few studies of prolotherapy to determine whether it is an effective therapy for rotator cuff tears [75]. Given the limited evidence, we do not recommend prolotherapy for the treatment of rotator cuff tears. (See "Overview of the management of overuse (persistent) tendinopathy", section on 'Prolotherapy'.)

●Stem cell therapy — Mesenchymal stem cells are an investigational treatment for rotator cuff tears and other musculoskeletal injuries purported to increase angiogenesis, promote healing, and alter the inflammatory cascade [76-79]. We do not support the use of this therapy for rotator cuff tears due to the dearth of high-quality evidence and the high cost. The use of stem cells for such treatment is reviewed separately. (See "Biologic therapies for tendon and muscle injury", section on 'Biologic therapies for tendon injuries'.)

●Topical nitroglycerin — Topical nitroglycerin has been used as an adjunct to physical therapy in the treatment of rotator cuff tears. There is evidence to suggest improvement in patients with chronic tendinopathy when treated with nitroglycerin [80]. (See "Overview of the management of overuse (persistent) tendinopathy", section on 'Topical nitroglycerin (glyceryl trinitrate)'.)

MONITORING PARTIAL-THICKNESS AND SMALL FULL-THICKNESS TEARS — We re-evaluate patients with partial-thickness and small full-thickness tears not being treated surgically, whether asymptomatic or mildly symptomatic, at regular intervals to determine whether symptoms or clinical signs of weakness or dysfunction have progressed and surgical referral is warranted. New or worsening pain or increased weakness or disability is often associated with an increase in tear size and a greater likelihood of requiring surgery. (See 'Indications for surgical referral' above.)

Initially, we monitor these tears at six weeks and again at three months while the patient is undergoing rehabilitation, and thereafter at 6- to 12-month intervals [39]. Some clinicians may choose to monitor patients thought to be at high risk for worsening tears and those with extreme frailty at more frequent intervals. After the initial 12 months, patients who regain good shoulder function and are pain free may follow up on an as-needed basis if any symptoms recur or function declines. For others, we continue surveillance until indications for surgical referral develop or the patient is either no longer interested in surgery or not a surgical candidate. In addition to regular monitoring, any new or significant increase in pain or any reduction in rotator cuff strength or function at any point typically warrants further imaging with MRI or ultrasound and possible surgical referral [81-83].

At follow-up visits for monitoring purposes, advanced imaging is not necessary unless the clinical assessment suggests that function has deteriorated, although clinicians skilled at ultrasound often perform an examination as part of their assessment. For clinicians using ultrasound, in our experience (while evidence is limited), tears that retract 1 cm or less are unlikely to progress further, tears that retract 1 to 2 cm are at some risk of progression, and tears that progress more than 2 cm generally require surgery.

At each follow-up visit for monitoring, a careful reassessment of symptoms and shoulder function should be performed. In addition to basic assessments of shoulder mobility and rotator cuff strength, including resisted external rotation, we perform the active painful arc test (picture 1) and drop arm test to assess function. (See "Physical examination of the shoulder", section on 'Examination for rotator cuff pathology'.)

Progression of rotator cuff tears is a common indication for surgical referral. Observational studies suggest that a significant percentage of rotator cuff tears increase in size over time, and that large increases are often (but not always) associated with new or increased pain and weakness [11,81,83-88]. As rotator cuff tears enlarge, the risk of concomitant muscle atrophy and tendon retraction also increases, which in turn increases the difficulty of and may preclude surgical repair. (See 'Prognosis' below.)

PROGNOSIS — The prognosis of rotator cuff tears varies widely depending on the injury and patient. Outcomes for tears managed nonoperatively are affected by the type and size of the tear, the effectiveness of physical therapy, and patient age, activity, and comorbidities [17,19,30,39,60,84]. Outcomes following surgery are discussed above. (See 'Surgical outcomes' above.)

Evidence about the prognosis of rotator cuff tears managed nonoperatively is limited. In a systematic review and meta-analysis of six trials comparing the outcomes for surgical and nonsurgical treatment, the prognosis was similar [89]. Although short-term outcomes for pain and function were better in the surgery group, the difference was not considered clinically significant. Longer-term outcomes are less well studied. What evidence we have for each tear type is summarized below.

Larger tears, whether traumatic or degenerative, often progress over time. Observational evidence suggests that greater than 80 percent of symptomatic, full-thickness tears and approximately 25 percent of partial-thickness tears increase in size [39]. The degree and rate of enlargement vary. Enlargement can occur over varying timeframes ranging from six months to nine years; thus, surveillance is warranted for surgical candidates. The rate of enlargement is faster in larger partial tears and full-thickness tears and appears to increase after the initial 12 months of follow-up. This increase in rate is more significant with full-thickness tears. Any new or worsening pain or dysfunction is often associated with an increase in tear size and a greater likelihood of requiring surgery [90]. (See 'Monitoring partial-thickness and small full-thickness tears' above and 'Indications for surgical referral' above.)

Many patients with smaller, less severe tears do well with physical therapy and are able to resume their previous activities, although overhead athletes are often unable to regain their preinjury level of performance. Such tears commonly become asymptomatic and may heal with scar tissue or, in some cases, normal-appearing tendon tissue. (See 'Overhead athletes' above.)

●Acute full-thickness tear – Limited evidence suggests the prognosis for nonoperative treatment of full-thickness tears treated successfully with physical therapy is comparable to that for surgically treated injuries, although the latter group improves slightly faster. Improvements seen with physical therapy are generally maintained beyond two years according to a prospective observational study of 595 patients with full-thickness tears [49]. However, in a randomized trial of 103 patients with small or medium-sized full-thickness tears, patients with surgically repaired injuries had less pain and better function at 10-year follow-up [18].

●Acute partial-thickness tear – The prognosis for acute partial-thickness tears varies with the size of the tear and the patient factors noted above. Larger acute tears are at greater risk of enlarging further and causing disability, ultimately warranting surgical referral, while smaller tears are more likely to respond well to physical therapy and remain stable [61]. In a systematic review of four observational studies involving 257 tears, the rate of progression to a full-thickness tear was approximately 0.26 percent per month (average follow-up 34 months) following nonoperative management [91].

●Acute-on-chronic tear – The prognosis for acute-on-chronic tears varies depending on the size of the acute tear, patient age, and other factors (eg, baseline function, fatty infiltration of tissue, comorbidities). Should a significant functional deficit develop, surgical referral is generally warranted. (See 'Indications for surgical referral' above.)

●Chronic degenerative tear (older adults) – The short-term prognosis for chronic degenerative tears is fairly good. Often, such tears respond to physical therapy, activity modification, and analgesia, and patients are able to perform their activities of daily living without undue difficulty. Observational evidence suggests that many older adults have degenerative tears but remain asymptomatic.

Nevertheless, there is a risk of chronic tears enlarging over time. Some tears may become symptomatic or cause disability, and some may be surgically irreparable. Larger tear size and worse atrophy are associated with poorer outcomes [84]. Atraumatic, full-thickness (ie, degenerative) tears that will respond to conservative management, including physical therapy, typically do so in the first 6 to 12 weeks. This finding is supported by a prospective observational study of 452 patients with degenerative tears managed with an evidence-based physical therapy program and close follow-up in which slightly fewer than 25 percent of patients ultimately opted for surgery [62]. Most patients who chose surgery did so between weeks 6 and 12 of the study.

RETURN TO SPORT AND WORK — The timeframe and capacity to return to sport and heavy labor following a rotator cuff tear depends on many factors, including the extent and type of injury, response to treatment, and the demands placed on the shoulder by the sport or work in question. For patients with relatively sedentary desk jobs, only one week of rest may be necessary, while patients with jobs involving heavy overhead lifting or those who participate in strenuous overhead sports may require six months or longer.

Patients who respond well to physical therapy may return when symptoms abate and they are able to perform the required tasks. Following surgical repair, the timeframe for return to activity depends upon the procedure performed. Arthroscopic debridement may have a significantly shorter time (as early as six weeks) compared with rotator cuff repair (typically five months) or shoulder arthroplasty, which may require one to five months or possibly longer [92-94]. A systematic review of 11 studies involving 385 athletes reported a wide range of rates (50 to 100 percent) and timeframes (4.8 to 14 months) for return to sport, thereby highlighting both limitations in the evidence and the highly contingent nature of recovery from arthroscopic surgery, which depends on the extent of injury, patient characteristics (eg, age), sport, and surgical procedure [95]. (See 'Surgical procedures and postoperative timeline' above.)

Among laborers treated with rotator cuff repair, over 35 percent were unable to return to their preinjury level of function at eight-month follow-up in systematic review and meta-analysis [96]. Rates were relatively higher among those with more physically demanding work, and the finding was independent of workers' compensation status. A retrospective study of 70 patients with workers' compensation claims reported similar results with over one-third unable to return to work at preinjury levels [97]. In this study, patients with three tendon tears experienced poorer outcomes.

SUMMARY AND RECOMMENDATIONS

●Rotator cuff tear types and presentation – Rotator cuff tears are characterized by timing (acute, subacute, or chronic), cause (trauma or chronic deterioration or combination), and extent (partial or full thickness; small, medium, large, or massive). Symptoms range from asymptomatic with preserved shoulder function to severe pain and substantial loss of shoulder function and strength. Tears and symptoms may progress over time. The presentation, examination, and diagnosis of rotator cuff tears are discussed in detail separately. (See "Presentation and diagnosis of rotator cuff tears" and 'Definitions and tear types' above.)

●Management by tear type – Little high-quality evidence is available to inform treatment decisions. Surgical repair may be necessary, especially for large, acute tears in active adults; physical therapy is the mainstay of nonoperative treatment. Surgical outcomes vary depending upon the characteristics of the tear and patient. (See 'Management and evidence limitations' above and 'Surgical management' above and 'Nonoperative management' above.)

Initial management for common types of tears is as follows:

•Partial-thickness tear – Most partial-thickness tears merit a trial of physical therapy regardless of patient or tear characteristics. Exceptions are individuals with severe weakness or functional deficits and those needing to return to a physically demanding sport, activity, or work. They should be referred to a shoulder surgeon. (See 'Acute traumatic tear' above.)

•Full-thickness tear (see 'Acute traumatic tear' above):

-Small (<1 cm) full-thickness tears have a slower rate of progression and may remain stable. Many can be treated with physical therapy. Patients whose function and symptoms do not improve with three months or more of appropriate physical therapy or worsen after the initial six weeks of physical therapy are referred to a shoulder surgeon.

-Larger (>1 to 1.5 cm) full-thickness tears have a high rate of tear progression. Patients with such injuries should be referred to a shoulder surgeon. Full-thickness tears with significant retraction and those that are large or massive in individuals who are suitable surgical candidates are generally treated operatively.

•Monitoring of patients managed with rehabilitation – We re-evaluate patients with partial-thickness and small full-thickness tears (regardless of symptoms) being treated with physical therapy at regular intervals to determine whether weakness or dysfunction has progressed and surgical referral is warranted. Initially, we monitor these tears at six weeks and again at three months while the patient is undergoing rehabilitation, and thereafter every 6 to 12 months. New or worsening pain or increased weakness or disability is often associated with an increase in tear size and need for referral. (See 'Monitoring partial-thickness and small full-thickness tears' above.)

•Massive tear – Massive tears involve extensive injury often involving more than one tendon and may cause severe disability. Patients with such tears should be referred to a shoulder surgeon to determine whether repair is feasible. (See 'Massive tear' above and 'Surgical management' above.)

•Chronic degenerative tear – Degenerative tears are more common among patients of advanced age and with multiple comorbidities (eg, diabetes). Treatment consists of pain relief, physical therapy, and activity modification. These injuries are generally not amenable to surgical repair. (See 'Chronic degenerative tear' above.)

●Overhead athletes – Overhead athletes often sustain partial-thickness rotator cuff tears from repetitive microtrauma and shoulder impingement. These tears can progress to full-thickness tears. Even for many younger athletes, physical therapy is the preferred initial approach to treatment since surgical outcomes are not ideal and many athletes are unable to resume their preinjury level of performance following surgery. (See 'Overhead athletes' above and 'Surgical outcomes' above and 'Return to sport and work' above.)

●Indications for surgical referral – Rotator cuff injuries that warrant immediate surgical referral include the following (see 'Indications for surgical referral' above):

•Acute, full-thickness, traumatic tear, especially in younger (<40 years), otherwise healthy patients.

•Sudden or severe weakness (eg, inability to abduct or flex shoulder) in a patient with a pre-existing partial-thickness tear. These findings are consistent with the development of a full-thickness tear.

•Failure to improve after three months of well-designed and properly performed physical therapy, or symptoms or function that worsen during the initial six weeks of such physical therapy.

In addition, we commonly make an early referral for specific patient populations so they can discuss the relative risks and benefits of repair with a shoulder surgeon, even if surgery is not the best option. These populations include:

•Heavy laborer with partial tear who depends on a functional shoulder for their livelihood

•Elite or professional overhead athlete

●Adjunct therapies – Multiple adjunct therapies have been used to treat rotator cuff tears, but few are supported by high-quality evidence. (See 'Unproven and investigational adjunct therapies' above.)

•Glucocorticoid injection – Subacromial glucocorticoid injection is commonly given to treat pain or inflammation associated with a tear. We believe it is reasonable to do so for patients whose pain prevents them from participating in physical therapy. We do not give a glucocorticoid injection to patients if a significant rotator cuff tear that may require surgery is known or suspected. A single injection given a minimum of three months prior to any surgical intervention is reasonable for patients requiring pain relief. (See 'Glucocorticoid injection' above.)

●Return to work and sport – The timing for return to work or sport is highly variable and dependent upon patient characteristics (eg, age), extent of tearing, treatment required, and demands the work or sport in question place on the shoulder. Rates of return to sport range from 50 to 100 percent and timeframes for recovery from 5 to 14 months. (See 'Return to sport and work' above.)