A variety of coexisting pathologic conditions will influence the decision-making regarding the treatment of proximal humerus fractures, including the need for prosthetic replacement of the proximal humerus. Osteoporotic bone is common in elderly patients with low-energy proximal humerus fractures and often compromises the ability to obtain rigid internal fixation of the fracture fragments. Preexisting glenohumeral arthritis may necessitate glenoid resurfacing if prosthetic replacement of the proximal humerus is undertaken. Preexisting rotator cuff tears are often seen in the elderly and may have significant implications with regard to final strength and range of motion of the shoulder. Degenerative lesions or acute disruption of the long head of the biceps tendon may require tenodesis to the proximal humerus following fracture reconstruction.

The classification of proximal humerus fractures is based on Codman’s identification of four major fracture fragments: the anatomic head, lesser tuberosity, greater tuberosity, and humeral shaft (14) (Fig. 12-3). Expanding on the work of Codman, Neer introduced his classification system for proximal humerus fracture-dislocations based on his observation of 300 fractures over a 14-year period (15). This is the most widely used classification system for proximal humerus fractures used today. His system takes into account fracture anatomy and fragment displacement, two factors that dictate choice of treatment options and provide valuable information about the prognosis of the fracture (10).

According to the Neer classification system, a fracture fragment is considered present only when it is displaced from the remaining proximal humerus or other fracture fragments. Displacement is noted when the fragment is separated by 1 cm or more or there is 45 degrees of angulation of the fragment. The clinical implication of displacement suggests potential instability of the fracture or possible disruption of the vascular supply to the humeral head. The majority of proximal humerus fractures are minimally displaced (15). Nondisplaced fracture lines or fragments with less than the previously mentioned displacement are classified as one-part injuries. Two-part fractures involve the greater tuberosity, lesser tuberosity, or surgical neck. Three- and four-part fractures involve a variety of combinations of the possible fracture fragments and are associated with higher rates of humeral head avascular necrosis. Neer also described proximal humerus fracture-dislocations that can occur with two-, three-, and four-part fractures. Dislocations are classified according to the direction of dislocation of the head fragment (anterior or posterior). In addition, Neer reported articular surface injuries including
head splitting and impression fractures. Both of these types of injuries can involve varying amounts of the humeral head with obvious implications regarding the viability of the articular surface (Fig. 12-4).

The AO group has proposed a thorough system of classification of proximal humerus fractures based on the precise location of the fracture fragments, direction and severity of fracture displacement, and associated dislocation of the articular surface (16). Fractures are initially group into isolated extraarticular unifocal (Type A), extraarticular bifocal (Type B), and articular isolation (Type C) fractures. The groups are then further classified based on fracture alignment, direction and amount of displacement, and the presence of head impaction or dislocation (18). The emphasis of this system is on the vascular supply of the humeral head and the subsequent risk of osteonecrosis. Avascular necrosis of the humeral head is uncommon after Type A and B fractures but frequently complicates Type C fractures. Although more descriptive than the Neer classification system, the AO classification of proximal humerus fractures has not gained widespread popularity for clinical application because of its complexity and limited reproducibility.

Proximal humerus fractures represent approximately 4% to 5% of all skeletal fractures (17). The majority of recent epidemiologic literature notes an increasing incidence of proximal humerus fractures that is felt to be related to osteoporosis in the aging population (17, 18, 19). Studies from Rochester, Minnesota, and Malmo, Sweden, have shown the age-adjusted incidence of proximal humerus fractures in adults to be 104-105 per 100,000 person-years (18,20). Proximal humerus fractures represent more than 75% of all humerus fractures in those older than age 40 years, with an incidence of approximately 70% of the reported rate of proximal femur fractures (18). The relationship of proximal humerus fractures to osteoporosis is confirmed by the correlation of a greater proportion of humerus fractures occurring proximally with advancing age and the age-related increase in osteoporosis. Additionally, there is a rapid increase in the incidence of proximal humerus fractures seen in postmenopausal women reflecting the common pattern of osteoporosis-related fractures (18,21,22).The incidence of proximal humerus fractures is likely to increase as the average lifespan of the population continues to grow, further emphasizing the need for effective treatments of these injuries to maintain the functional independence of affected individuals.

The majority of proximal humerus fractures occur as the result of a fall. In younger patients with good bone stock, high-energy mechanisms are common and the fractures are often accompanied by other injuries. In multi-trauma victims, proximal humerus injuries are often overlooked because of the presence of more severe musculoskeletal or other life-threatening injuries. Older patients typically sustain proximal humerus fractures as a result of minimal to moderate trauma, such as a fall from standing height. Proximal humerus fractures have also characteristically been associated with seizure disorders or electric shock (23,24).

Associated injuries to the brachial plexus and axillary artery can accompany proximal humerus fractures (25,26). Stableforth reported a 4.9% incidence of axillary artery
injury in displaced proximal humerus fractures (27). Neer’s series of displaced proximal humerus fractures had no associated vascular injuries (15). The most common site of injury is proximal to the branch of the anterior humeral circumflex artery (28). Arterial injury, although rare, has also been reported even with minimally displaced fractures (29).

Clinically detectable nerve injuries following proximal humerus fractures have been reported in up to 45% of cases (30). Stableforth reported a 6.1% incidence of brachial plexus injuries following displaced proximal humerus fractures (27). Isolated injuries to the axillary nerve have been reported as well (31). A recent electromyography study revealed a very high incidence of occult nerve injury following both nondisplaced (52%) and displaced (82%) fractures (32). The axillary and suprascapular nerves were most commonly involved. Often these injuries are incomplete and may manifest as temporary weakness that recovers along the same course of time as fracture healing. The risk of neurovascular injury following fractures of the proximal humerus is greater in high-energy injuries, fracture-dislocations, and penetrating trauma.

Although rare, concomitant chest-wall injuries have been reported with high-energy proximal humerus fractures (15,27). These include rib fractures, pneumothorax, and/or hemothorax. There have been several reports of intrathoracic humeral head penetration following displaced surgical neck fractures (33, 34, 35).

The majority of proximal humerus fractures occur secondary to acute events well recognized by the patient. A thorough history can provide valuable information regarding the mechanism of injury and the likelihood of associated injuries. Younger patients typically sustain higher energy injuries such as motor vehicle accidents or falls from heights. These fractures can result from direct blows to the shoulder or more commonly from indirect mechanisms such as a fall to the outstretched hand. A high index of suspicion during the history can help the clinician identify symptoms of a more serious injury such as upper-extremity paralysis or paresthesias, open wounds, or dyspnea.

Older patients typically have lower energy injury mechanisms. Falls are the most frequent cause of fracture in this age group (21). A history of frequent falls or amnesia to the events surrounding the fall may signal cardiac or central neurologic problems. A history of head trauma or prior seizure disorder should be elicited as well. Altered sensorium noted during examination after a fall warrants immediate neurologic evaluation and imaging to rule out acute injury to the brain.

The physical examination of the shoulder following proximal humerus fractures can be quite variable reflecting the wide spectrum of injury seen with this fracture. The majority of patients will complain of local pain in the upper arm and shoulder areas. Soft-tissue swelling occurs early, whereas ecchymosis is usually seen after 24 to 48 hours in gravity-dependent areas. Fracture-dislocations can alter the normal soft-tissue contours of the upper extremity. There is usually generalized tenderness to palpation of the proximal humerus that may be more localized with isolated fractures of the tuberosities. Attempted motion of the shoulder is painful and may be associated with crepitus and muscle guarding. The entire upper extremity, neck, and chest wall should also be examined for evidence of associated injuries.

A careful neurovascular examination is critical in the initial evaluation, especially with high-energy fractures. Timely identification and treatment of vascular injuries is critical in avoiding the complications of a dysvascular limb. The radial pulse is palpated and compared to the opposite arm when possible. Because of the extensive collateral circulation of the upper arm, a radial pulse may be palpated in the presence of a significant arterial injury (28). Doppler ultrasonography or angiography can be used to further assess the vascular anatomy in cases with diminished or absent pulses. Diffuse distal paresthesias are a sign of inadequate circulation and should raise the suspicion of vascular injury (28). The neurologic examination should encompass all components of the brachial plexus. Axillary nerve integrity is assessed by both sensory and motor function, although the latter is often inhibited by pain. Atony of the deltoid after proximal humerus fracture creating inferior subluxation of the humeral head is common and usually recovers in the subacute period (36, 37, 38). Distal motor and sensory examination of all peripheral nerves allows the clinician to assess each component of the brachial plexus and should be repeated at regular intervals if abnormalities are discovered.

Radiographic evaluation of proximal humerus fractures is the foundation of proper patient evaluation and treatment. Plain radiography is the gold-standard method of initial evaluation and will serve to direct appropriate treatment. The standard series obtained to evaluate these injuries has been coined the “trauma series” and includes the anteroposterior (AP) and lateral views in the plane of the scapula and the axillary view. Alternatively, the authors recommend a standard AP view of the shoulder, an AP view of the scapula, and an axillary view. These orthogonal views allow multiplanar assessment of the fracture allowing appreciation of fracture lines and fragment displacement as well as orientation of the humeral head articular surface in relation to the glenoid (Fig. 12-5).

The scapula rests obliquely on the chest wall angled medially (toward the opposite shoulder) 35 to 40 degrees from the coronal plane. To obtain the AP and scapular lateral views, the angle of the beam and cassette should match the orientation of the scapula. The scapular AP view is taken with the cassette along the posterior aspect of the scapula and the beam angled perpendicular to the cassette. This view shows the true AP view of the glenohumeral joint with no superimposition of the humeral head and glenoid in the normal setting. The scapular lateral (scapular
Y) view is taken with the cassette along the anterolateral shoulder with the beam directed from posterior and medial toward lateral along the axis of the scapular spine. Displacement of tuberosity and shaft fragments and humeral head to glenoid orientation can be appreciated on this view. Both the AP and scapular lateral views may be obtained while the patient remains in a sling and can be performed in a variety of positions if needed. The standard AP view of the shoulder is taken with the cassette placed along the posterior shoulder parallel to the thorax while the beam is directed from anterior to posterior perpendicular to the thorax. Because of the orientation of the scapula on the chest wall, this image creates an oblique view of the glenohumeral joint. Because the arm usually rests in a position of internal rotation against the trunk, this provides an excellent view of the greater tuberosity. The axillary view provides valuable information that may not be seen on the AP and lateral views. Although this view is obtained with the glenohumeral joint abducted, only a small degree of elevation is necessary for an adequate radiograph and can usually be obtained in the acute setting. Physician assistance is sometimes required for proper handling of the acutely injured extremity. The cassette is placed superior to the shoulder, and the beam is directed to the cassette through the axilla. The axillary view best shows the relationship of the humeral head to the glenoid as well as posterior displacement of greater tuberosity fractures, anteriorly displaced lesser tuberosity fractures, head impression injuries, and associated glenoid rim fractures.