Forward flexion of the shoulder is defined as elevation of the humerus in front of the body in the sagittal plane (Fig. 2.1). This motion is typically governed by contraction of the anterior fibers of the deltoid muscle and weakness is often a key indicator for several different pathologies. Although the position of the elbow joint was never specified by the AMA [3] or AAOS [4] back in the 1960s, most practitioners measure flexion range of motion with the elbow extended. However, there are some other maneuvers that can be used specifically to measure deltoid strength. For example, having the patient make a fist and push anteriorly against the examiners hand would also activate the deltoid and simulate a forward elevation motion without fully elevating the shoulder overhead (see Chap. 3). This method is particularly useful when evaluating deltoid strength either before or after an arthroplasty procedure where full elevation may not be possible.

Abduction of the shoulder occurs when the humerus is elevated in the coronal plane such that the extremity points directly laterally (also known as straight lateral abduction) (Fig. 2.2). This position should be differentiated from abduction within the scapular plane which places the humerus in approximately 20–30° of forward angulation, also termed “scaption” (Fig. 2.3). This slight forward angulation facilitates examination such that surrounding soft tissues are similarly lax on both the anterior and posterior sides of the joint. For example, examination of the patient with straight lateral abduction of the shoulder would place an increased stress on anterior structures relative to posterior structures (Fig. 2.4). Thus, any evaluation of the capsular structures may produce inaccurate results when the humerus is abducted in the coronal plane. For this reason, appropriate glenohumeral and scapulothoracic resting positions should be utilized such that accurate assessment can be achieved (the scapular plane and the glenohumeral resting positions are discussed in more detail below).

Extension of the shoulder typically refers to any position in which the humerus rotates beyond the scapular plane (Fig. 2.5). This can be achieved with the humerus at the side or elevated. With the arm at the side, the humerus can extend posteriorly in a limited capacity. Straight lateral abduction (also known as “horizontal abduction”) as discussed above is also considered a position of extension since the humerus would be angulated posterior to the plane of the scapula. Similarly, the humerus can also extend posteriorly when the arm is overhead. For example, throwing athletes require combined overhead extension and external rotation to achieve maximal torque and potential energy. Sometimes, this motion sequence can lead to pathologic problems such as symptomatic internal impingement, SLAP tears, and scapular dyskinesis.

Internal rotation describes motion around a center of rotation such that the angular motion vector points towards the midline. For example, when viewing a right shoulder from the superior to inferior direction, a counterclockwise rotation of the humerus would be referred to as internal rotation (Fig. 2.6). Internal rotation with the arm in a position of 90° of abduction or 90° of flexion uses the same concept. The humerus rotates in the same direction in each case, regardless of the position of the elbow, forearm, or hand. Internal rotation can also be measured with the arm in the adducted position. In this case, the patient will attempt to reach as far up the spinal column as possible while the clinician determines the most superior spinal level that the patient can reach (Fig. 2.7). This position maximizes internal rotation and has historically been a standard measure for internal rotation capacity. However, this method of measurement has recently been called to question since the vertebral level to which one reaches may be influenced by elbow, wrist, and hand motion rather than isolated internal rotation of the humerus [5].

When viewing a right shoulder from superiorly to inferiorly, external rotation would be defined as clockwise rotation of the humerus away from the midline (see Fig. 2.6). Again, similar to internal rotation, the rotational moment about the humeral anatomic axis does not change whether the humerus is abducted or flexed—it is only the scapulohumeral angle that changes.

Shoulder adduction can also be described with the arm at the side or elevated. The basic resting position with the arm at the side is often referred to as “simple adduction.” When the humerus is elevated to 90° followed by movement of the humerus towards the opposite shoulder, this is most often referred to as “horizontal adduction” or “cross-body adduction” (Fig. 2.8). Conversely, “horizontal extension” corresponds to the opposite motion, where the humerus is extended posteriorly beyond the scapular plane.

The scapular plane is generally defined as a position of neutral scapulohumeral angulation which optimizes glenohumeral joint congruity and facilitates accurate and consistent evaluation of the joint. This position of neutral scapulohumeral angulation is determined by the angle between a line drawn along the center axis of the scapula and second line drawn at the same level that is perpendicular to the coronal plane (see Fig. 2.3). This position, which most often occurs between 20° and 30° of forward angulation relative to the coronal plane with the humerus in various degrees of abduction, minimizes the potential for acromiohumeral contact while also allowing for the theoretical isolation of the rotator cuff musculature during various clinical examination tests. In other words, some have theorized that abduction of the humerus within the scapular plane requires zero contribution from internal or external rotators to achieve full abduction capacity [6]. Maximum capsuloligamentous laxity also occurs within the scapular plane (at the glenohumeral resting position, discussed below) which facilitates examination of these structures (instability and laxity testing are discussed in Chap. 6).

Although the scapular plane is generally defined as 20–30° of humeral forward angulation relative to the coronal plane in normal individuals, it must be recognized that patients with scapular malposition or dyskinesis, as which occurs commonly in overhead athletes, may have a scapular plane that differs from the rest of the population. For example, a throwing athlete with scapular malposition may display increased protraction and upward rotation in the resting position (discussed below), thus altering the position of the glenoid such that the plane of the scapula occurs with greater forward angulation of the humerus. Therefore, performing physical examination tests within the “normal” scapular plane in a patient with scapular malposition may produce inaccurate results (specific examination maneuvers for evaluation of the scapulothoracic articulation are presented in Chap. 9).

Also known as the “loose pack position,” the resting position of a joint is the position at which surrounding soft tissues are under the least amount of tension, the joint capsule has its greatest laxity and the bony surfaces of the joint are minimally congruent [7–9]. In other words, this position is considered to allow maximal glenohumeral mobility owing to an increase in joint laxity [8]. The glenohumeral resting position in normal shoulders is thought to be between 55° and 70° of abduction with the humerus in neutral rotation within the plane of the scapula (Fig. 2.9) [10–12]. In this position, the amount of external force required to translate the humeral head is minimal which is thought to facilitate examination accuracy. Although there is a general consensus regarding the location of the glenohumeral resting position, validation studies have seldom been conducted.

In a cadaveric study, An et al. [13] evaluated arm elevation in positions of either internal or external rotation. In this study, maximum elevation occurred with the arm externally rotated within the plane of the scapula. They could not achieve this maximal elevation with the arm internally rotated due to the acromiohumeral impingement that occurs in this position. In other words, there was bony contact between the acromion and the greater tuberosity, thus hindering the ability to further elevate the arm. When the humerus was placed in a position of 30° of forward angulation, there was little contribution from the internal and external rotators during humeral abduction.

In 2002, Hsu et al. [14] used seven cadaveric specimens to measure the translational and rotational range of motion at different angles of humeral abduction within the scapular plane. The glenohumeral resting position was calculated as the mid-point of the confidence intervals where maximal rotational and translational motion occurred. Maximal anteroposterior translation and maximal rotational range of motion occurred at approximately 39° of humeral abduction in the scapular plane and corresponded to approximately 45 % of the maximum available abduction range of motion. They also found that the glenohumeral resting position varied according to the maximal available range of motion, possibly suggesting that patients with joint hypermobility and hypomobility should be tested at greater and lesser degrees of humeral abduction, respectively. Since this was a cadaveric study, the effect of dynamic glenohumeral stabilization (which also contributes to the resting position) could not be evaluated.

More recently, Lin et al. [9] attempted to define the glenohumeral resting position in vivo in the dominant shoulders of 15 healthy patients. In that study, translational and rotational range of motion capacities were determined using an electromagnetic tracking device after an 80 N translational load and a 4 N-m (torque) rotational load were applied. The greatest maximal rotational range of motion occurred at approximately 49.8° of abduction in the scapular plane. However, in contrast to Hsu et al. [14], the greatest maximal anterior–posterior translation occurred at approximately 23.7° of abduction in the scapular plane. These results suggested that testing for anteroposterior joint laxity should be conducted at lower degrees of abduction than when testing for rotational joint laxity.

Considered together, these studies demonstrate the complexity and potential variability that the glenohumeral resting position can have across a population, between populations or even between individuals (dominant versus non-dominant shoulders). In general, it is important to determine the maximal range of translational and rotational range of motion for each patient. In general, the rotational resting position is thought to occur at a point near 45 % of the total abduction arc [14] where half of this abduction angle is thought to represent the translational resting position [9].

Codman’s paradox is the observation that as the arm is flexed upward in the sagittal plane and let down in the coronal plane, the humerus appears to rotate 180° as evidenced by the orientation of the palm. In other words, when beginning the motion, the palm faces posteriorly and, at the end of the motion, the palm faces anteriorly (Fig. 2.10). Alternatively, an individual can place their hand at the top of the head through either (1) forward flexion and internal rotation or (2) abduction and external rotation. This observation has traditionally been of academic interest; however, many investigators have attempted to mathematically solve the “paradox” using complex equations and algorithms [15, 16]. Although the clinical relevance of Codman’s paradox is debatable, some authors have investigated an application of Codman’s paradox during manipulation of a stiff shoulder under anesthesia [17]. In addition, the quadrant test (discussed later in this chapter) is based on Codman’s paradox and can be a useful measure of global shoulder motion [18].

With the arm at rest, the scapula is predictably positioned in a specific orientation that can be used to detect scapular malposition before any motion measurements or evaluations are undertaken. There have only been a few studies that quantified the precise location of the scapula on the posterior thorax. Sobush et al. [25] quantified the normal scapular resting position in cadavers using the “Lennie test,” or a series of measurements taken from the superomedial and inferomedial angles of the scapula. The distance from the superomedial angle to the midline, the distance from the inferomedial angle to the midline and also the angle of scapular inclination were determined (i.e., the angle formed between a line connecting the spinous processes and a line drawn along the margin of the medial scapular border). This test was found to have high inter-rater reliability and accuracy when compared to post-measurement radiographs. Using similar measurements, a cadaveric study by Fung et al. [26] found that the resting position of the scapula was at approximately 3° of external rotation, 40° of internal rotation, and 2° of posterior tilt. Of note, this nomenclature does not reflect the position of the humerus. Rather, it represents the position of the scapular body relative to the coronal plane (discussed below).

In order to evaluate three-dimensional scapulothoracic motion, it is perhaps most advantageous to begin with an understanding of the basic two-dimensional scapular motions. In total, there are three rotational movements and two translational movements (Fig. 2.11). Although it is not possible to isolate these movements, they represent the basic components that comprise three-dimensional scapular motion.

Internal and external rotation occurs around the vertical axis of the scapula—that is, internal rotation elevates the medial scapular border away from the posterior thorax (i.e., the glenoid faces more anteriorly) whereas external rotation refers to the exact opposite motion (i.e., the glenoid faces less anteriorly).

Upward and downward rotation occurs along the plane of the scapula. In other words, upward rotation occurs when the inferior angle of the scapula moves laterally and the glenoid faces more superiorly. Conversely, downward rotation refers to the opposite motion in which the inferior scapular angle moves medially towards the midline and the glenoid faces more inferiorly.

Anterior and posterior rotation (i.e., tilting) occurs around the horizontal axis of the scapula. Anterior tilting of the scapula occurs when the inferior angle moves away from the thorax (and the superior border moves towards the thorax) whereas posterior tilting refers to the exact opposite motion in which the inferior angle moves towards the thorax (and the superior border moves away from the thorax).

The scapula can also translate in the medial–lateral direction (as in protraction and retraction, described below) and the superior–inferior direction (as in shrugging the shoulders). It is important to recognize that these translational motions also require intact AC and SC joints—upward and downward translation of the scapula requires upward and downward angulation of the clavicle via the SC joint whereas medial–lateral translation requires anterior–posterior motion of the clavicle through the SC joint as the scapula moves around the thorax.

Three-dimensional scapular motion, which is achieved by combining any of the above-mentioned two-dimensional movements, is necessary to optimize glenohumeral contact and stability throughout the entire range of shoulder motion. However, during the evaluation of an actual patient, it is most useful to consider the observed three-dimensional scapular motion as a summation of the individual rotational moments mentioned above.

The terms “protraction” and “retraction” are most often used to describe scapular movement in three-dimensional space. To understand these terms, it is perhaps easiest to first recognize that scapular motion occurs along a rounded surface (i.e., the convexity of the posterior thorax). Using this approach, one could imagine that any lateral translation of the scapular body would also require scapular internal rotation. This movement also requires some anterior tilt and downward rotation. This combination of movements is generally referred to as scapular “protraction” and can be closely simulated by having the patient thrust their shoulders anteriorly (similar to a hunchback position). Conversely, any medial translation of the scapular body would also require scapular external rotation. This movement also requires some posterior tilt and upward rotation. This combination of movements is typically referred to as scapular “retraction” which can be demonstrated by having the patient thrust their shoulders posteriorly (as in “squeezing” the scapulae together by extending the humerus posteriorly below 90° of elevation) (Fig. 2.12). Of course, neither protraction nor retraction could be achieved without some amount of upward and downward rotation along with anterior and posterior tilt; however, the purpose of the above example is to illustrate the fundamental concept of scapular translation around the posterior chest wall in three dimensions.

This same concept also applies when the scapula translates superiorly or inferiorly along the convex surface of the posterior thorax. In other words, inferior translation of the scapula would theoretically produce an increased posterior tilt whereas superior translation of the scapula would produce an increased anterior tilt (as in shrugging the shoulders). In reality, the shoulder shrug requires a combination of superior translation, anterior tilt, and internal rotation. Increased anterior or posterior scapular tilt is very subtle, is difficult to recognize by direct visual or tactile examination in the office setting, and generally cannot be isolated by any specific voluntary movement. Biomechanical studies suggest that scapular tilting mostly occurs during extension-type maneuvers with the arm either overhead or at the side.

Most clinicians agree that isolated glenohumeral motion occurs below approximately 90° of elevation whereas combined glenohumeral and scapulothoracic motion occurs above this level (discussed below). In order to maximize glenohumeral contact and stability during this combined motion, the scapular stabilizers must not only contract in synchrony with each other, but also with each of the muscles that cross the glenohumeral joint along with the proprioceptive feedback obtained from surrounding soft-tissue structures (such as the glenohumeral joint capsule [6]). Although a thorough discussion of each possible scapular movement is beyond the scope of this book, we aim to emphasize the extreme importance of understanding the fundamental concepts related to three-dimensional scapular motion. It is with this foundational knowledge that one can begin to understand the complex disease processes related to the shoulder.

The clavicle acts as a strut which allows for the strategic positioning of the shoulder girdle along the side of the thorax. In order to maximize shoulder range of motion, the clavicle must be dynamically positioned according to scapular motion via the AC and SC joints. Therefore, the health of the clavicle and the AC and SC joints is extremely important to achieve normal scapular motion in three-dimensional space [26–28]. For example, arm elevation requires the clavicle to retract, elevate, translate, and rotate posteriorly along its long axis (i.e., the so-called “screw axis”) where each of these movements is dependent on the function of intact, painless AC and SC joints [29, 30].

Unfortunately, visual estimation is the most commonly used method for the measurement of shoulder range of motion. Although several studies have found that experienced practitioners can estimate range of motion with a similar accuracy to standardized measurement devices [41, 42], the lack of teaching regarding the fundamentals of range of motion testing is disappointing. This results in inaccurate measurement estimations by the novice examiner who was never properly taught to use goniometers or inclinometers. Although formal measurement requires more time to complete, it is suggested that inexperienced examiners use standard measurement devices to aid in accurate patient assessment until they become more knowledgeable and experienced with the examination process. In a busy clinical practice, however, the experienced clinician can usually make rapid range of motion estimations without sacrificing accuracy or precision. Visual inspection and estimation of range of motion is therefore a standard of practice in most cases, but formal measurements are required when a study involving range of motion data is being conducted.

An inclinometer is essentially a leveling device, similar to that which is used by a carpenter to measure the degree of inclination of a surface relative to the horizontal plane, that is occasionally used in clinical practice to quantify shoulder range of motion or, in some cases, the degree of spinal deformity such as kyphosis or scoliosis [43–45]. Both mechanical and digital inclinometers have been described as reliable and valid tools for the measurement of shoulder range of motion.

Mechanical inclinometers, or hygrometers, use gravity and a fluid-level indicator to measure the inclination of the humerus relative to the horizontal plane in degrees (Fig. 2.14). The first reported use of a mechanical inclinometer to measure range of motion was in 1975 by Clarke et al. [31]. In their study, the inter-observer reliability was approximately 0.93. Similarly, Dover et al. [46] calculated inter- and intra-rater reliability values of approximately 0.99 in the measurement of shoulder range of motion using a mechanical inclinometer. In an adjunct study [47], the same group found that the ability of the inclinometer to detect changes in joint proprioception was also pronounced. In that study, they calculated inter- and intra-observer reliabilities ranging from 0.978 to 0.984.

Currently, digital inclinometers are more commonly used to assess shoulder range of motion [48–57] and function by calculating the angle of inclination relative to the horizontal plane using an implanted gravity sensor. Kolber et al. [51] determined that the inter- and intra-observer reliabilities of digital inclinometry was greater than 0.95 and found that this method was interchangeable with goniometry (intra-class correlation [ICC] coefficients for goniometry were >0.94) [51]. Scibek and Carcia [54] studied 13 healthy collegiate subjects in an attempt to quantify scapulohumeral rhythm using a digital inclinometer. In that study, the investigators found that the scapula contributed to 2.5 % of shoulder motion in the first 30° of shoulder abduction. This proportion dramatically increased to between 20.9 and 37.5 % when measured between 30° and 90° of humeral abduction, respectively. Johnson et al. [58] calculated the reliability and validity of a digital inclinometer to measure scapular upward rotation during humeral abduction in the scapular plane. They found that the digital inclinometer had excellent reliability and validity in the assessment of scapular motion with inter- and intra-observer ICCs ranging from 0.89 to 0.96. A similar study by Tucker and Ingram [56] calculated ICCs of >0.89 after using a digital inclinometer to quantify scapular upward rotation with static humeral elevation.

More recently, studies by Shin et al. [59] and Mitchell et al. [60] demonstrated the ability of smart phone inclinometers (and goniometers) to accurately measure range of motion with excellent inter- and intra-observer reliability (ICC >0.9). One other study [61] demonstrated the capability of smart phones to measure cervical range of motion. This method of measurement eliminates the cost of standard digital inclinometers, a factor that has limited their widespread use. Nevertheless, these studies demonstrate the utility and practicality of digital inclinometers in the accurate measurement of scapulohumeral rhythm in addition to glenohumeral range of motion capacity.