The coracoacromial ligament originates from the distal-lateral extension of the coracoid process and travels posterosuperiorly to insert upon the anterolateral margin of the acromion [15, 16]. As part of the coracoacromial arch, this ligament is commonly described as being involved with rotator cuff impingement lesions due to the proximity of the cuff tendons that pass closely beneath, especially as the arm is elevated. The coracoacromial ligament has a number of anatomic variations [17–19]; however, only those variations that involve a distinct anterolateral and posteromedial bundle are likely to be related to rotator cuff impingement and subsequent tearing. In a cadaveric study, Fremery et al. [20] found that shoulders with rotator cuff tears and clinical evidence of impingement had stronger, thicker anterolateral bands compared to shoulders with unrelated issues. Evidence of traction spur formation within the anterolateral band has also been found which further implicates its involvement with the development of impingement [21]. Chambler et al. [22] suggested that arm abduction results in increased tension of the coracoacromial ligament which may provide an explanation for the development of traction spurs in these patients. A more recent cadaveric study by Yamamoto et al. [23] found that the superior cuff made contact with and, in fact, generated increased tension through the coracoacromial ligament during range of motion testing in a series of normal, healthy shoulders. This finding may provide at least one possible explanation behind the development of traction-type spurs on the anterolateral acromion with advancing age, potentially leading to extrinsic compression of the superior cuff tendons. However, whether or not the thickness of the anterolateral band is a cause or effect of rotator cuff disease has not been elucidated.

The acromion is also subject to developmental abnormalities as a result of failed fusion of secondary ossification centers. This failed fusion results in a defect known as an “os acromiale” and occurs in approximately 8 % of the population where 1/3 of these individuals are affected bilaterally [24]. Os acromiale is a mobile accessory ossicle that, when unstable and pulled inferiorly by contraction of the deltoid with arm elevation, has been associated with the development of identifiable impingement lesions and pain at the top of the shoulder (Fig. 4.7). However, this relationship has been refuted on several occasions [25–29]. In addition, surgical treatment strategies for os acromiale that involve increasing the volume of the subacromial space has not resulted in an improvement in clinical outcomes [26]. Further study is therefore needed to clarify the effects of os acromiale on normal rotator cuff tendons.

The anterior aspect of the acromion may, in itself, be a potential site of rotator cuff abrasion and subsequent tearing regardless of the presence or absence of space-occupying traction spurs associated with the coracoacromial ligament. In 1991, Bigliani et al. [30] described the three most common acromial morphologies as flat (type I), curved (type II), and hooked (type III), citing the hooked acromion as most relevant to the development of subacromial impingement (Fig. 4.8). In a cadaveric study, Flatow et al. [13] demonstrated that the type III acromion has an increased propensity to make contact with the rotator cuff tendons when compared to the other acromial morphologies that were previously described by Bigliani et al. [30]. Wang et al. [31] found that acromial morphology influenced the success of conservative management for rotator cuff tears. In their study, the majority of patients with either a type I or II acromion responded favorably to conservative management. In contrast, more than half of those patients with a type III acromion failed nonoperative treatment and required subsequent surgical intervention. Unfortunately, these authors did not report their findings at the time of surgery. In another study, Gill et al. [14] found that acromial morphology was an independent factor associated with the development of rotator cuff pathology using multivariate logistic regression analysis. Additionally, Natsis et al. [32] found a statistically significant increase in the rate of anterolateral acromial spur formation in those with a type III acromion. They concluded that a type III acromion with anterolateral spur formation was contributing factor associated with the development of rotator cuff impingement and tearing. In 2012, Hamid et al. [33] arrived at similar conclusions. However, despite these results, there are also a number of studies that refute the oft-cited correlation between the type III acromion and the development or progression of rotator cuff tears [17, 31, 34–38]. As a result of this conflicting data, further study is needed to determine if acromial morphology, as described by Bigliani et al [30], is truly associated with the development of symptomatic subacromial impingement and rotator cuff tears.

More recently, other acromial morphologies have been also been described—namely, the convex acromion (type IV; as described by Vanarthos and Monu [39] as an addendum to the original classification system developed by Bigliani et al. [30] Fig. 4.9) and the keeled acromion (originally described by Tucker and Snyder [41] Fig. 4.10). While the type IV acromion has not been implicated as an anatomic factor associated with rotator cuff disease, the keeled acromion may be involved to some extent. The keeled acromion, which was described as a central spur (or convexity) located on the undersurface of the acromion, was significantly associated with the presence of both partial- (bursal-sided) and full-thickness rotator cuff tears in the original study published by Tucker and Snyder [41]. Several recent studies have also suggested that a steep acromial slope may be another factor associated with the development of impingement lesions [34, 42–44]; however, further study needs to be conducted to substantiate these claims.

Excessive lateral extension of the acromion, which is best quantified through calculation of the acromial index (Fig. 4.11), has also been reported as a potential contributor to the development of rotator cuff impingement and tearing. Some investigators report that decreased coverage of the humeral head by the acromion (i.e., a decreased acromial index) may allow the humeral head to utilize the anterolateral acromion as a fulcrum or lever to aid in glenohumeral elevation, possibly causing abrasion of the cuff tendons and subsequent rotator cuff tearing [45, 46]. Nyffeler et al. [47] suggested that a large acromial index may result in a more superiorly directed force vector produced by the middle fibers of the deltoid, potentially leading to superior migration of the humeral head which, in turn, decreases the available space for the cuff tendons to pass beneath the acromion. This theory has been partially validated since other more recent studies have found statistically significant associations between increased acromial indices and the presence of rotator cuff tears [31, 33, 34, 47–50]. Although the acromial index appears to play some role in the development of rotator cuff disease, additional studies are needed to fully elucidate the exact pathomechanisms behind this phenomenon.

Inclination of the glenoid in the coronal plane has also been associated with the development of rotator cuff tears on several occasions [51–53]. Tétrault et al. [52] performed measurements using magnetic resonance imaging (MRI) to determine the orientation of the supraspinatus muscle fibers relative to the glenoid surface. In that study, the mean angle formed between the supraspinatus tendon and the glenoid surface was approximately 80° in the coronal plane. The authors suggested that a decrease in this angle (i.e., increased upward tilt of the glenoid) may result in a more vertically oriented force vector produced by the supraspinatus, thus preferentially pulling the humeral head superiorly [52]. If this theoretical mechanism is factually correct, the supraspinatus tendon could then make contact with the acromion, possibly leading to the cascade of events commonly associated with rotator cuff disease. A similar mechanism may occur when considering glenoid anteversion and retroversion in which tearing of the subscapularis and infraspinatus is observed, respectively [52]. Although at least one study found that surgically decreasing the glenoid inclination angle may decrease the measured amount of superior humeral head translation with passive abduction [49], none of the more recent imaging studies have shown significant associations between any type or degree of glenoid version and the presence of rotator cuff lesions, regardless of location of the tear or the tendon involved [54, 55].

The critical shoulder angle (CSA) (recently described by Gerber et al. [56] and Moor et al. [57, 58]) is another radiographic measurement purported to have an association with rotator cuff tears or osteoarthritis. The CSA is obtained from true anteroposterior (AP) radiographs and takes into account both the acromial index and the degree of glenoid inclination in the scapular plane (Fig. 4.12). The original developers of this measurement have suggested that the CSA may have an ability to predict the future development of degenerative rotator cuff tears (when the CSA > 35°) and glenohumeral osteoarthritis (when the CSA < 30°). However, a well-designed prospective study would be needed to confirm these claims given the current lack of conclusive clinical data suggesting any association between either the acromial index or glenoid inclination and any shoulder pathology.

The Neer impingement sign, first described by Neer [10] in 1972, is elicited by passive and maximal forward elevation of the humerus and stabilization of the scapula with the examiner’s contralateral hand (Fig. 4.14). Stabilization of the scapula is essential to maximize the utility of the test since upward rotation of the scapula (and therefore the acromion) with forward elevation will decrease the likelihood of reproducing cuff impingement under the acromion. Reproduction of the patient’s symptoms is indicative of a positive test. Several investigators have evaluated the clinical efficacy of the Neer impingement sign in its ability to accurately diagnose subacromial impingement (Table 4.1) [72, 73, 75, 76, 78, 80].

Both Fodor et al. [73] and Kelly et al. [79] used ultrasonic evaluation to determine the sensitivity and specificity of the Neer sign in the diagnosis of subacromial impingement. Interestingly, although each study reported similar sensitivity values, their specificity values were divergent (95 % and 10 %, respectively). These results highlight the significant variability that may exist in the performance and interpretation of physical examination findings, specifically with regard to subacromial impingement syndrome. Nevertheless, Hegedus et al. [81] attempted to account for various confounding factors and study quality in a recent meta-analysis. In that study, the overall calculated sensitivity of the Neer impingement sign was 72 % while its overall specificity was approximately 60 %. Clearly, this maneuver is not adequate to diagnose subacromial impingement in isolation; however, combination of its results with those obtained from the Hawkins–Kennedy test and the painful arc sign (described below) are likely to improve diagnostic accuracy.

The Hawkins–Kennedy test [70] was first described in 1980; however, it was not originally thought to be as reliable or reproducible as the Neer impingement sign. A positive Hawkins–Kennedy test is the result of greater tuberosity contact on the undersurface of the coracoacromial ligament that is thought to reproduce symptoms related to subacromial impingement. To perform this test, the shoulder is brought to 90° of abduction in the scapular plane with the elbow also flexed 90°. From this position, the humerus is slowly and maximally internally rotated (Fig. 4.15). Reproduction of the patient’s symptoms (typically pain over the anterior shoulder) is deemed a positive test and may be indicative of superior cuff impingement. Numerous studies have evaluated the efficacy of the Hawkins–Kennedy test in its ability to diagnose subacromial impingement with variable results (Table 4.2) [72–76, 78, 80, 82]. The meta-analysis by Hegedus et al. [81] reported similar overall sensitivities and specificities for both the Neer impingement sign and the Hawkins–Kennedy tests.

The painful arc sign [71] is elicited with resisted abduction of the shoulder in the scapular plane (20–30° of forward angulation) with the elbow in full extension (Fig. 4.16). Pain with this maneuver is another possible indicator of subacromial impingement, especially when used in combination with the tests described above. Although this test is not very sensitive in isolation, it does boast a modest specificity as reported by Hegedus et al. [81] (Table 4.3). Therefore, since both the Neer impingement sign and the Hawkins–Kennedy test each has low specificity values, addition of the painful arc sign (which has a higher reported specificity) may improve overall diagnostic accuracy when all three maneuvers are used in a composite exam to diagnose subacromial impingement.