Presentation and diagnosis of rotator cuff tears

INTRODUCTION — Shoulder pain is a common presenting complaint to primary care offices and sports medicine clinics. Estimates of the prevalence of shoulder pain range from 16 to 34 percent in the general population [1,2]. Rotator cuff pathology is the most common condition of the shoulder for which patients seek treatment [3]. Estimates published in 2004 by the United States Agency for Health Care Research and Quality and United States Department of Health and Human Services show rotator cuff problems accounted for more than 4.5 million clinician visits and approximately 40,000 surgeries [4].

Despite this burden on patients and health care resources, there remains significant controversy regarding the etiology of rotator cuff injuries, the role of rotator cuff tears in generating shoulder pain, and the ability of commonly accepted clinical tests to diagnose rotor cuff tears.

The etiology, presentation, and diagnosis of rotator cuff tears will be reviewed here. The treatment of rotator cuff tears, rotator cuff tendinopathy, the shoulder examination, and the general management of patients with shoulder pain are discussed separately. (See "Management of rotator cuff tears" and "Rotator cuff tendinopathy" and "Physical examination of the shoulder" and "Evaluation of the adult with shoulder complaints".)

ANATOMY AND BIOMECHANICS — The anatomy and biomechanics of the shoulder generally are discussed separately (figure 1A-C) (see "Evaluation of the adult with shoulder complaints", section on 'Anatomy and biomechanics'). The biomechanics of the rotator cuff specifically are discussed elsewhere. (See "Rotator cuff tendinopathy", section on 'Clinical anatomy' and "Rotator cuff tendinopathy", section on 'Basic biomechanics'.)

MECHANISM OF INJURY — The development of underlying rotator cuff tendinopathy is discussed separately. (See "Rotator cuff tendinopathy".)

The cause of rotator cuff tears is likely multifactorial. Degeneration, impingement, and overload may all contribute in varying degrees to the development of rotator cuff tears.

Several theories have developed to explain the cause of rotator cuff injury. In 1934, Codman theorized that rotator cuff tears developed from intrinsic tissue degeneration [3]. This theory was bolstered by Rothman and Parke, who suggested in 1965 that a "critical zone" of hypovascularity in the rotator cuff predisposed the tendon to degeneration. The role of intrinsic factors was challenged in 1972 by Neer, who suggested that impingement from extrinsic structures caused rotator cuff tears [3].

Subsequent investigations in throwing athletes suggest that tension overload contributes to these injuries [5,6]. Tension overload is thought to occur when the ability of the rotator cuff to compress and maintain the stability of the glenohumeral joint is overwhelmed by the distractive forces of throwing or by trauma. In other words, rotator cuff muscle weakness allows subluxation of the glenohumeral joint, leading to impingement, which contributes to the development of rotator cuff tears.

Most often, rotator cuff lesions begin as partial tears of the undersurface or articular portion of the supraspinatus tendon (figure 2) [4]. Over time they can progress to full thickness tears to include the supraspinatus, infraspinatus, subscapularis and biceps tendons.

RISK FACTORS — Rotator cuff tears are primarily a disease of middle aged and older patients. Observational data reveals a nearly linear increase in the frequency of rotator cuff tears over time [4,7-14]. However, a sizeable portion of tears in older patients are asymptomatic. (See 'Symptoms' below.)

Sports and occupations requiring overhead activity produce a high frequency of rotator cuff tears. In these patients, tears may present at a younger age and are frequently associated with labral pathology [3,5].

A number of medical conditions have been associated with rotator cuff tears, but whether they contribute directly to such injuries or are confounders is unclear. A meta-analysis of 26 studies involving more than 3000 patients reported an association between symptomatic rotator cuff tears and diabetes, hypertension, higher body mass index (BMI), and smoking [15].

Individual studies report myriad findings. A longitudinal cohort study confirmed an association with older age and higher BMI [15,16]. A similar study reported elevated risk among diabetics that persisted despite insulin therapy but was reduced in non-diabetic patients with hypercholesterolemia treated with statin therapy [17]. In an observational study of 180 subjects with a symptomatic rotator cuff tear, an asymptomatic contralateral rotator cuff tear was identified with ultrasound in 69 individuals [18]. The prevalence of an asymptomatic lesion was greater among individuals with diabetes, obesity, hypercholesterolemia, and hypertension and among those who smoked or engaged in heavy repetitive labor involving the shoulder. Another cohort study reported an increased probability of rotator cuff tear in patients with cervical spine disease [19].

Genetics and familial predisposition may play a role in rotator cuff pathology. Several haplotypes have been implicated [20]. It remains unclear if genetics exerts its effects through anatomy that predisposes to injury (eg, distance between acromion and humeral head), intrinsic tendon weakness, or some combination of factors.

Trauma can cause rotator cuff tears. One study found that 58 percent of patients presenting to the emergency department with acute shoulder trauma and normal radiographs who were unable to abduct above 90 degrees had acute traumatic tears of the rotator cuff [12]. Tears were diagnosed by ultrasound during follow-up evaluations performed a median of 13 days following the injury.

CLINICAL PRESENTATION

Symptoms — Pain and weakness may be the presenting complaints of the patient with rotator cuff tear. Classically, rotator cuff tears are associated with pain [21,22]. Pain is said to develop over the lateral deltoid and is exacerbated by overhead activities and at night (possibly when the patients lies on the shoulder).

This classic description may be inaccurate. Several observational studies suggest that pain associated with rotator cuff tears, when present, is nonspecific and may be due to associated subacromial bursitis, not the tear itself [7,23]. Other observational studies suggest a large portion of rotator cuff tears are asymptomatic [7-9,24]. Although some degeneration of the rotator cuff may be inevitable in most individuals beyond the fifth decade of life, these degenerative changes do not necessarily cause significant functional impairment.

To confuse matters further, some researchers find that increasing tear size correlates with more severe symptoms [7,10], while others do not [25]. Some researchers maintain that partial thickness tears create significantly more pain and disability than full thickness tears [26,27].

The understanding of the relationship between rotator cuff tears and shoulder pain is evolving. Clinicians should not allow the presence or absence of pain to play too significant a role in diagnosis.

Weakness is another frequently cited symptom of rotator cuff tears and in large or full thickness tears represents an integral part of the diagnosis. Weakness may be apparent in otherwise asymptomatic individuals [28-30]. However, using the physical examination to isolate rotator cuff weakness can be problematic. In addition, pain from bursitis, impingement, or trauma can lead to reflex muscle inhibition and weakness unrelated to rotator cuff injury. Distinguishing between reflex muscle inhibition and true weakness is the rationale behind using a diagnostic injection of lidocaine as part of the clinical evaluation for rotator cuff tears. (See "Rotator cuff tendinopathy".)

Physical examination — We use a combination of three tests to determine the likelihood of a rotator cuff tear and the need for further workup:

●Active painful arc test (picture 1)

●Drop arm test (failure to smoothly control shoulder adduction)

●Weakness in external rotation (picture 2)

Performance of these and other examination techniques for the shoulder is discussed separately. (See "Physical examination of the shoulder", section on 'Examination for rotator cuff pathology'.)

Although many textbooks and clinicians have long advocated for using the physical examination to determine the presence of rotator cuff pathology, well-performed studies to support particular examination techniques or approaches are scarce. Variations in methodology among studies are partially to blame. Definitions of a positive test vary. Some use pain, others use weakness, and still others use both. Diagnostic gold standards also vary and may include advanced imaging or arthroscopic and surgical findings.

Our approach to the examination is based primarily upon our clinical experience and the results of a prospective observational study of 552 patients who were systematically evaluated using eight common shoulder examination tests and then underwent diagnostic arthroscopy [31]. Logistic regression analysis found that the combination of a positive painful arc sign, positive drop arm test, and weakness in external rotation accurately predicted full-thickness rotator cuff tear (LR 15.6), while a combination of three negative tests made the diagnosis unlikely (LR 0.16). This study is among the few well-performed trials to assess shoulder examination techniques that include a sufficient number of patients to draw meaningful conclusions [32].

A similar study assessed the accuracy of 23 clinical tests in 400 patients who subsequently underwent arthroscopic surgery [11]. According to this study, patients with shoulder pain who demonstrated supraspinatus and infraspinatus weakness and a positive impingement sign had a 98 percent chance of having a full thickness tear. Among patients 60 years or older with positive findings of any two of these three tests, there was again a 98 percent chance of tear. The study did not include descriptions of the techniques used for clinical testing, making it difficult to replicate the protocol. We have included descriptions of the most commonly used tests for the signs they describe (picture 2 and picture 3 and picture 4 and picture 5).

A prospective observational study of 208 patients with shoulder pain reported that the “Hornblower Sign,” which involves resisted external rotation while the shoulder is held in 90 degrees of abduction, is specific for infraspinatus tear, but additional study is needed [33].

If one assumes a pretest probability for rotator cuff tear of 30 to 70 percent, the combination of clinical tests described in either of these two large, prospective studies would effectively rule in full thickness tears if all were positive. If all were negative, the latter study's combination would rule out full thickness rotator cuff tear [11] while the former study's combination would rule out full thickness tear in patients over 60 years of age [31].

Systematic reviews of studies assessing clinical tests for the shoulder report a wide range of approaches and methodologic quality, thereby precluding meaningful meta-analysis of tests for rotator cuff integrity [34,35]. In addition, individual analyses of the 15 available studies in one review showed that only 8 studies met quality criteria and none of the individual clinical tests included among these studies was consistently diagnostic [34]. This highlights the difficulty in recommending evidence-based clinical tests for rotator cuff tears and the importance of continued research. The results of another systematic review of five relatively high quality studies of maneuvers to detect rotator cuff injury reinforces the limitations of studies performed to date and the potential for both type 1 and type 2 errors when clinicians attempt to detect rotator cuff tears based upon a single maneuver [35]. Accurate diagnosis of partial thickness tears or tears of specific rotator cuff muscles and tendons through clinical testing remains challenging, and our review of the literature did not yield sufficient high quality evidence to determine whether any particular test or combination of tests is best for diagnosing these conditions. We believe that rotator cuff tears should be suspected or ruled out on the basis of the history and a combination of examination tests, and prefer the combination listed at the beginning of this section.

As studies of rotator cuff examination maneuvers accumulate, two consistent findings are noteworthy:

●No single examination maneuver can accurately diagnose a rotator cuff tear.

●Several different combinations of examination tests can be useful for diagnosis.

Alternative test combinations advocated by clinical researchers include the following:

●The empty can, full can, and zero-degree abduction tests to diagnose supraspinatus tears; the drop sign and resisted external rotation test to diagnose infraspinatus tears [36,37].

●The Jobe and full can tests to diagnose supraspinatus tears; the lag and Hornblower tests to diagnose infraspinatus tears [33]. In this study, 15 tests were assessed using expert examination and magnetic resonance imaging (MRI) as the diagnostic gold standard.

●In a select group of patients presenting with a history suggesting rotator cuff tear but negative results for traditional examination maneuvers, the Dynamic Isotonic Manipulation examination (DIME) was used to determine the need for MRI [38]. A negative DIME test, following negative traditional testing, effectively ruled out rotator cuff tear and made MRI unnecessary.

The concept of muscle lag provides the basis for several published tests of rotator cuff integrity. These tests require further study and are discussed separately. (See "Physical examination of the shoulder", section on 'Testing for rotator cuff tear'.)

We do not routinely use injections of an anesthetic (eg, lidocaine) specifically to help diagnosis rotator cuff tear. Selective anesthetic injection at discrete locations (eg, acromioclavicular joint, proximal biceps tendon) can help to determine the cause of shoulder pain in some cases. Subacromial injection of an anesthetic and glucocorticoid can help to relieve symptoms in some patients. The use of such injections is discussed separately. (See "Rotator cuff tendinopathy", section on 'Adjunct treatments' and "Acromioclavicular joint disorders" and "Biceps tendinopathy and tendon rupture", section on 'Glucocorticoid injection for tendinopathy' and "Frozen shoulder (adhesive capsulitis)", section on 'Glucocorticoid injection'.)

IMAGING STUDIES

Plain radiographs — Although usually normal, anterior-posterior, lateral, and outlet views of the shoulder can be of value in assessing large, chronic rotator cuff tears. These studies can help to confirm the diagnosis by showing whether the humeral head is migrating relative to the glenoid and acromion. The clinician can expect larger rotator cuff tears (>1.75 cm) and probable involvement of both the supraspinatus and the infraspinatus among individuals with a symptomatic shoulder and humeral migration on plain radiographs [39]. Plain films also allow for the evaluation of some concomitant or alternative shoulder pathology.

Musculoskeletal ultrasound — Musculoskeletal ultrasound (MSK US) is an accurate tool for the evaluation of superficial tendon and muscle lesions, as well as bursitis, of the shoulder, and enables dynamic examination at the bedside. Its role in the evaluation of the rotator cuff tendinopathy and performance of the shoulder ultrasound examination are discussed separately. (See "Rotator cuff tendinopathy", section on 'Musculoskeletal ultrasound' and "Musculoskeletal ultrasound of the shoulder".)

When performed by skilled operators, musculoskeletal ultrasound demonstrates diagnostic accuracy for supraspinatus tears comparable to MRI. In a meta-analysis of 23 studies involving 2054 shoulders assessed by ultrasound and using arthroscopy as the reference standard, median sensitivity and specificity for full thickness supraspinatus tears were 88 and 93 percent, respectively [40]. For partial thickness supraspinatus tears, test characteristics were less impressive, with sensitivity 65 and specificity 86 percent. When assessed by year of publication, more recent studies showed better results, possibly reflecting improved technology and training.

Magnetic resonance imaging (MRI) — MRI accurately diagnoses full thickness rotator cuff tears and many partial tears (figure 2), although its sensitivity for detecting partial tears is much lower (image 1) [41-44]. The accuracy of MRI for detecting tears may also be limited in patients with severe glenohumeral osteoarthritis [45]. MRI often provides information about the degree of tear, tendon retraction, and muscle atrophy, all of which is critical in preoperative planning for rotator cuff repair. Conversely, MRI may reveal tears that are not amenable to operative management, such as chronic retracted tendon tears. (See "Management of rotator cuff tears".)

The MRI "cuff" protocol, designed specifically to identify supraspinatus pathology, may improve sensitivity particularly for the non-radiologist, but further study is needed [46]. According to a meta-analysis of 14 studies involving imaging of over 1800 shoulders, MRI is sensitive and specific for identifying full-thickness subscapularis tendon tears but has only modest sensitivity and specificity for partial-thickness subscapularis tears [47].

Interpretation of the MRI must be clinically correlated due to the high rate of asymptomatic tears. An observational study of MRI performed in 96 asymptomatic individuals showed rotator cuff tears in 34 percent of subjects, and nearly 54 percent of those over 60 years of age [24]. MRI can be used if the diagnosis remains unclear after initial evaluation or if definitive confirmation of the diagnosis is needed for guidance about surgery or return to sport.

Rotator cuff tears are diagnosed on MRI based upon discontinuity of the tendon on T-1 weighted images and a fluid signal on T-2 weighted images. Ancillary findings include fluid in the subacromial space on T-2 images, loss of the subacromial fat plane on T-1 images, and proliferative spur formation of the acromion or acromioclavicular joint. Large chronic tears may be associated with cephalad migration of the humeral head and fatty atrophy of the supraspinatus and infraspinatus muscle bellies [48]. In addition to rotator cuff pathology, MRI allows visualization of periarticular soft tissues, including the capsulolabral complex and biceps tendon.

Arthrography with MRI or CT — Magnetic resonance arthrography (MRI-A) is more sensitive and specific for diagnosing partial thickness rotator cuff tears than standard MRI, but it too has limited accuracy, especially for certain types of tears [41,42,49-52]. According to a meta-analysis of 12 studies, MRI-A is as accurate as MRI for identifying bursal-sided, partial-thickness tears [53]. MRI-A is not typically obtained as part of the diagnostic workup for rotator cuff tears and is usually reserved for cases when a labral injury is suspected.

According to small prospective studies using findings at arthroscopy as the gold standard, the diagnostic performance of arthrography with computed tomography (CT) is similar to that of MRI-A for detecting rotator cuff tears [50]. CT arthrography may be useful in patients requiring more sophisticated imaging to assess the rotator cuff but who cannot undergo MRI-A (eg, body metal, pacemaker). The role of arthrography with MR or CT to assess the shoulder is reviewed separately. (See "Radiologic evaluation of the painful shoulder in adults", section on 'Arthrography'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Shoulder soft tissue injuries (including rotator cuff)".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Rotator cuff injury (The Basics)")

●Beyond the Basics topic (see "Patient education: Rotator cuff tendinitis and tear (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Mechanism and risk factors – Rotator cuff tears are common, particularly in older adults. (See 'Mechanism of injury' above and 'Risk factors' above.)

The etiology of rotator cuff tears is multifactorial. Chronic tears likely represent the culmination of degeneration and impingement. Trauma can produce acute tears. (See 'Mechanism of injury' above.)

●Clinical presentation – Many rotator cuff tears are asymptomatic. Patients with symptoms may complain of shoulder pain, weakness, and difficulty with overhead or reaching activities. (See 'Clinical presentation' above.)

●Physical examination – The use of three clinical tests (painful arc, drop arm sign, and weakness on external rotation) can accurately diagnose full thickness rotator cuff tears in patients older than 60 years of age. (See 'Physical examination' above.)

●Diagnostic imaging – The identification of partial tears of the rotator cuff and tears of individual rotator cuff tendons remains a clinical challenge and the use of clinically correlated imaging is needed to make the diagnosis. Musculoskeletal ultrasound and MRI can accurately make the diagnosis of most rotator cuff tears. (See 'Imaging studies' above.)