Additional videos related to the subject of this chapter are available from the Medizinische Hochschule Hannover collection. The following videos are included with this chapter and may be viewed at expertconsult.inkling.com :
Pipkin fractures—anterior approach.
Surgical hip dislocation in the treatment of Pipkin fractures.
Femoral Head Fractures
Femoral head fractures often occur in association with hip dislocations. Eighty-five percent to 90% of hip dislocations are posterior. In the largest series of posterior hip dislocations, the incidence of associated femoral head fracture was 7%. The common patterns of femoral head fracture are of the shear, or cleavage, type. The phenomenon of indentation or crush fracture has been more recently recognized, and the results in this group of patients seem to be worse than those in the cleavage group. These injuries have been reported to be commonly associated with anterior hip dislocation but are now frequently being recognized in association with acetabular fractures.
Of the 238 published cases identified by Brumback and coworkers, only 24 (10%) were associated with anterior hip dislocations. In a series of reported anterior hip dislocations, 15 of 22 (68%) had associated femoral head fractures. Because anterior dislocations occur infrequently, additional data on the association of femoral head fractures are lacking; however, indentation fracture of the femoral head seems to be commonly associated. The anatomic variations of a shallow acetabulum and femoral neck retroversion may play a role in predisposing patients to traumatic hip dislocation.
The anatomy of the proximal end of the femur, particularly the vascular anatomy, plays a critical role in determining outcome. The end results of fracture healing, fragment resorption, or femoral head avascular necrosis (AVN) are determined by the traumatic effect of the hip dislocation on the vascular anatomy. These results are influenced to some degree by management of the injury. Similarly, the effects of traumatic dislocation on femoral and acetabular articular cartilage can lead to arthrosis, which may be functionally limiting. Arthrosis, too, can be affected to some extent by management of the injury. Finally, damage to the hip capsule and hip musculature may lead to periarticular fibrosis and heterotopic ossification, which can produce functional limitations.
The femoral head is supplied by three terminal arterial sources: the artery of the ligamentum teres; a terminal branch of the lateral femoral circumflex artery; and the terminal branch of the medial femoral circumflex artery, the lateral epiphyseal artery ( Fig. 54-1 ). The last source is the critical blood supply to most of the weight-bearing superior portions of the femoral head. In most cases in which hip dislocation is associated with a femoral head fracture, the direction of the dislocation is posterior. The medial femoral circumflex artery is stretched, and the lateral epiphyseal artery may be occluded because of pressure from the edge of the disrupted posterior hip capsule and posterior acetabular wall. Intracapsular hematoma does not result because of loss of capsular integrity. The anteroinferior femoral head fragment generally remains within the acetabulum attached to the ligamentum teres. The intact blood supply to this fragment and the artery of the ligamentum teres from the obturator artery allow fracture healing to occur. The plane of the fracture, especially in posterior hip dislocation, most likely disrupts the osseous branches of the terminal divisions of the lateral femoral circumflex artery. The tension or occlusive pressure on the lateral epiphyseal artery makes it critical to promptly reduce the femoral head back into the acetabulum. AVN of the femoral head increases in frequency the longer the hip remains dislocated. This temporal effect is also true when the dislocation is associated with fracture of the femoral head.
Articular cartilage covers the proximal femoral epiphysis, which roughly involves the weight-bearing hemisphere. The cartilage reaches a maximal thickness of 4 mm in the superior-most region and tapers as it approaches the equator of the hemisphere. It thins in the region of the insertion of the ligamentum teres. At the periphery of the cartilage, the retinaculum vessels penetrate the bone. Approximately 70% of the entire articular surface of the femoral head is involved in load transfer. Damage to this surface, such as that produced by fracture of the femoral head, decreases the total surface of the femoral head available for load transfer. Accompanying increases in peak compressive forces may lead to breakdown of the articular cartilage matrix, loss of the articular seal, and the development of posttraumatic osteoarthritis. Femoral head indentation fractures are associated with acetabular fractures and anterior hip dislocations. They produce the same effects with focal crushing of cartilage matrix and loss of total contact area.
The adult human femoral head ranges in diameter from 40 to 60 mm and is not a perfect sphere. Out-of-round estimates are in the 1- to 1.5-mm range. This subtle asphericity is reflected on the acetabular side and was previously thought to be an important factor in prosthetic design. Accurate reduction of femoral head fragments that involve the articular cartilage is necessary to maximize contact between the femoral head and acetabulum. Reduction of femoral head fragments also minimizes peak stresses across the articular cartilage.
Maintenance of optimal femoral head–acetabular contact requires the entire femoral head. Loss of a significant piece of the femoral head would allow radial-lateral, noncongruent (shear) motion. How large the anteroinferior fragment has to be to compromise this “shim” effect is not known. The short-term clinical results of resection of small fragments have been satisfactory in some series and poor in others.
Mechanism of Injury
The vast majority of femoral head fractures are secondary to motor-vehicle crashes. The mechanism in most cases associated with posterior hip dislocation is similar to that believed to produce femoral neck, shaft, or combination fractures. The thigh is axially loaded on impact with the dashboard, and if the femoral shaft does not fracture, a hip injury will result if sufficient force is present. If the thigh is abducted, a femoral neck fracture may result, and if neutral or adducted, posterior hip dislocation with or without a concomitant femoral head or acetabular posterior wall fracture may result ( Fig. 54-2 ). Femoral head fractures may be the result of avulsion by the ligamentum teres or cleavage over the posterior margin of the acetabulum. Especially in anterior dislocations, impacted femoral head fractures may result from direct contact with the anterior wall of the acetabulum. In rare cases, femoral head fractures can be bilateral.
Consequences of Injury
Degenerative Joint Disease
Hip dislocations occur as a result of high-velocity injury. Significant force is required to disrupt the posterior hip capsule and more may be required to add a shearing injury and produce a femoral head fracture as the head is impaled on the posterior rim of the acetabulum. Crushed, indented, or fragmented articular cartilage results in loss of function of this critical tissue. If the injury is associated with poor reduction, loss of bone stock, or excision, the mechanical environment for the remaining articular cartilage is negatively affected, thereby adding further impetus to the breakdown of cartilage matrix. If significant posterior acetabular wall bone loss is also seen, posterior hip instability adds to the deterioration in hip function. In the same manner, loss of the medial shim effect produces a poor environment for survival of the remaining intact femoral head cartilage. The end result of the trauma and subsequent inferior conditions for articular cartilage is degenerative arthritis of the hip and poor hip function. Because most of these injuries occur in young adults, subsequent reconstruction becomes problematic. Total hip replacement has a higher rate of failure in the long term for this patient population. Hip arthrodesis, although effective in limiting pain and optimizing function, is not an attractive option for most patients. Malunion of the femoral head fractures can lead to pain and limited hip range of motion (ROM). Late excision or partial ostectomy of smaller femoral head fragments has been reported to have reasonable short-term results.
AVN is frequently reported in association with posterior hip dislocation. It accompanies 13% of posterior hip dislocations and is seen in 18% of such dislocations associated with femoral head fracture. The higher incidence may be a result of the greater amount of force required to produce the accompanying fracture, which also produces more soft tissue disruption. In addition, delay in closed reduction may occur because of the fracture surfaces or interposed fragments. Delay in closed reduction has been associated with higher rates of AVN after hip dislocation with fracture of the femoral head. Optimal management of hip dislocation is required to minimize the risk of AVN because in a young adult, this complication is a devastating problem without good options for treatment. AVN appears to be more commonly associated with posterior surgical approaches than with anterior surgical approaches.
Limited Motion (Heterotopic Ossification)
Poor functional results frequently occur after dislocation of the hip complicated by femoral head fracture. In addition to joint arthrosis and AVN of the femoral head, femoral head fracture is often associated with heterotopic ossification. Such ossification results from disruption of the joint capsule and contusion, tearing, and avulsion of the abductor musculature. Heterotopic ossification can also be associated with surgical exposure. Occasionally, a type I fracture can heal to the acetabulum, affecting hip motion.
The association of femoral head fractures with hip dislocation is strong. It is difficult to conceive how a shearing fracture of the femoral head could be produced without dislocation. However, patients can present with the hip not remaining dislocated because it spontaneously reduced. Indentation fractures frequently accompany acetabular fractures and result from “central dislocation” with impaction of the femoral head on the acetabular fragments. Management of hip dislocations can have an impact on the incidence of sciatic nerve palsy because delayed reduction of a hip dislocation results in an increasing chance of incidence and severity of sciatic neurapraxia. The axial loading mechanism described previously explains the not infrequent association of knee ligament injury, patella fractures, and femoral shaft fractures. The knee and femur must be carefully examined in patients with femoral head fractures because the force is usually transmitted through these structures.
Because these injuries are a result of high-energy trauma, injury to other body systems is frequent. Critical evaluation of the whole patient by the trauma team must be performed.
Diagnosis and Evaluation
Most femoral head fractures occur as a result of high-energy trauma such as motor-vehicle accidents, pedestrian versus motor vehicle, and falls from significant height. Although the mechanism of posterior dislocation is believed to be axial loading of a flexed and adducted hip and that of anterior dislocation to be abduction, flexion, and external rotation, most patients are unable to give such detailed descriptions. Information from the emergency medical service should assist with determining the mechanism and evaluating for associated injuries.
The associated hip dislocation, if it remains unreduced, determines the findings of the examination on admission. Posterior dislocation leaves the limb shortened, slightly flexed, adducted, and internally rotated. The anterior obturator type of dislocation results in the injured limb being flexed, abducted, and externally rotated. The position of the limb should be noted, and then rapid assessment of circulatory status should be performed, including pulses, capillary refill, and skin temperature. This examination must be followed by a thorough assessment of sciatic and femoral nerve function. The ability or lack thereof to dorsiflex and plantar flex the ankle, invert and evert the foot, and flex and extend the knee should be evaluated by palpating the muscle bellies as the indicated motion is attempted, followed by a careful sensory examination involving light touch and pinprick. No reduction of the hip joint should be attempted until this examination is complete and documented. It is important to examine the ipsilateral extremity, especially the knee (dashboard injury). Associated bony or ligamentous injury to the ipsilateral knee or femoral shaft is not uncommon. When associated with a femoral shaft fracture, a dislocation may go unrecognized because the classic position of flexion, internal rotation, and adduction is not apparent. Dedicated radiographs will be needed if suspicious of other lower extremity injury.
An anterior-posterior (AP) pelvic radiograph is a routine part of the evaluation of a multiply injured patient (see Chapter 9 ). If a routine chest, abdomen, and pelvis computed tomography (CT) scan has been performed, appropriate reformatting may provide much of the initial imaging required for evaluating a femoral head fracture. For patients with an isolated injury and suspected hip dislocation, proximal femoral fracture, or pelvic fracture, an AP radiograph must be obtained because the findings on this critical radiograph would determine which other radiographic studies are needed. In the case of a posterior hip dislocation, the radiograph must be scrutinized with regard to femoral head fragments remaining in the acetabular fossa. The femoral head defect is not obvious unless the angle of the radiographic beam catches the plane of the femoral head fracture in profile. In obvious posterior hip dislocation, the AP pelvis will show a superiorly displaced femoral head, a void in the acetabular socket, and a disruption of Shenton’s line compared to the opposite hip. To avoid displacement of an undisplaced femoral neck component of a Pipkin type III fracture, the femoral neck should be carefully scrutinized before any decision is made to reduce the hip. If the radiograph clearly demonstrates a hip dislocation with or without a concomitant femoral head fracture, the surgeon should proceed with a closed reduction maneuver. If the dislocation is associated with disruption of the anterior or posterior pelvic ring, pelvic inlet and outlet views should be obtained after reduction so as not to delay the reduction procedure. Similarly, if an associated acetabular fracture is suspected either on the contralateral side or as in a Pipkin type IV fracture, radiographic evaluation should include the 45-degree oblique views described by Judet and Letournel.
After obtaining the best possible plain radiographs, an attempt is usually made at closed reduction, which may be done in the emergency department with analgesia and conscious sedation or in the operating room with general anesthesia and complete muscle relaxation. Although the latter is probably less traumatic, it may not be an available option without excessive delay. Thus, it is often appropriate to attempt gentle closed reduction in the emergency department. If such reduction is unsuccessful and if further studies do not delay general anesthesia, CT through the acetabulum and femoral head at 2-mm cut intervals should be rapidly performed. If open reduction becomes necessary, a CT scan will assist the surgeon in searching for loose bodies and interposed soft tissue or in performing open reduction and internal fixation (ORIF) of an associated femoral head or acetabular fracture. It additionally serves as a valuable source of information for the surgeon regarding the choice of surgical approach. A CT-directed pelvic oblique radiograph can also be of assistance in accurately determining the size of the fragment, as well as any displacement.
If closed reduction is successful in either setting, it must be confirmed by a follow-up AP pelvic radiograph. Additional Judet views can be beneficial after reduction to confirm concentric reduction and evaluate for other injuries such as acetabular, femoral head, and neck fractures. A postreduction CT scan of the involved hip should be obtained to evaluate for hip joint congruency, loose bodies, posterior wall acetabular fracture, femoral head fracture pattern, and the amount of femoral head fracture displacement.
In certain settings, electromyography, contrast venography, cystography and urethrography, bone scans, and magnetic resonance imaging (MRI) may be useful as part of the prereduction or postreduction evaluation. Patients with hip dislocations (especially those that remain dislocated for long periods) may have associated sciatic nerve palsy. Electromyography can play an important role in determining the specific areas of the nerve that are involved and the degree of involvement. This information is helpful in relating the prognosis for recovery to the patient, especially if electromyograms are repeated serially. The initial study should not be obtained until 3 weeks after injury to allow accurate diagnosis. Duplex ultrasonography is a convenient and reliable test for deep venous thrombosis (DVT) in the thigh and popliteal fossa, but contrast venography or MRI angiography remain the confirmatory “gold standard” for proximal DVT. Urethrography and cystography are seldom indicated in pure hip dislocation with an associated femoral head fracture but may be indicated in cases with an associated anterior pelvic ring fracture and significant displacement (see Chapter 40 ). Technetium bone scanning can offer some predictive information regarding the chance of later AVN. If femoral head uptake is significantly lower than that of the contralateral normal hip as measured by quantitative scintimetry, the risk of later AVN may be as high as 80% to 90% but is dependent on multiple factors. Finally, MRI may offer some prognostic information in regard to the risk of femoral head AVN. The exact clinical implications of an abnormal femoral head MRI signal are yet to be clearly defined, but MRI can explicitly identify bone contusion, sciatic nerve contusion, osteochondral fracture, and fractures of the acetabular rim or femoral head; however, its routine use cannot be recommended until clear cost-benefit advantages are apparent. The accuracy of MRI in identifying intraarticular fragments seems to be much lower than that of CT.
The first recognition of femoral head fracture as a unique entity was published in 1869 by Birkett. Thompson and Epstein’s classification of posterior hip dislocations, published in 1951, included the classification of femoral head fracture as a separate entity ( Table 54-1 ). This classification did not include anterior hip dislocation, nor did it include fractures of both the acetabulum and femoral head.
|I||With or without minor fracture|
|II||With a large single fracture off the posterior acetabular rim|
|III||With comminuted fractures of the acetabular rim (with or without a major fragment)|
|IV||With fracture of the acetabular rim and floor|
|V||With fracture of the femoral head|
Stewart and Milford’s classification, published in 1954, did include the distinction between anterior and posterior hip dislocations. The associated fractures were classified as shown in Table 54-2 . Again, the system was limited by the inability to include fractures of the acetabulum with femoral head fractures. Additionally, classification of the acetabular component was lacking in detail. Because more conditions are clearly included, the Thompson-Epstein classification was used in most publications of the 1950s and 1960s.
|I||No acetabular fracture or only a minor chip|
|II||Posterior rim fracture, but stable after reduction|
|III||Posterior rim fracture with hip instability after reduction|
|IV||Dislocation accompanied by fracture of the femoral head or neck|
Pipkin’s landmark article on femoral head fractures included his new classification system ( Fig. 54-3 ). This article has remained the most significant contribution to the subject and it is the most commonly used classification for femoral head fracture. The Pipkin classification is shown in Table 54-3 . This classification describes the location of the femoral head fracture and whether there is associated femoral neck or acetabular fracture. The major deficiencies of this classification are the lack of differentiation of anterior hip dislocation and insufficient expansion of the acetabular fracture categorization. The last point is minor, and the need for the first was not apparent to Pipkin because the cases on which this classification was based were collected from his Kansas City associates and were probably all associated with posterior hip dislocations.
|I||Hip dislocation with fracture of the femoral head caudad to the fovea capitis femoris|
|II||Hip dislocation with fracture of the femoral head cephalad to the fovea capitis femoris|
|III||Type I or type II injury associated with fracture of the femoral neck|
|IV||Type I or type II injury associated with fracture of the acetabular rim|
The association of femoral head fracture with anterior hip dislocation has become more apparent in recent years, and Brumback and coworkers have published the most complete classification ( Table 54-4 ).
|1||Posterior hip dislocation with femoral head fracture involving the inferomedial, non–weight-bearing portion of the femoral head|
|1A||With minimal or no fracture of the acetabular rim and a stable hip joint after reduction|
|1B||With significant acetabular fracture and hip joint instability|
|2||Posterior hip dislocation with femoral head fracture involving the superomedial, weight-bearing portion of the femoral head|
|2A||With minimal or no fracture of the acetabular rim and a stable hip joint after reduction|
|2B||With significant acetabular fracture and hip joint instability|
|3||Dislocation of the hip (unspecified direction) with associated femoral neck fracture|
|3A||Without fracture of the femoral head|
|3B||With fracture of the femoral head|
|4||Anterior dislocation of the hip with fracture of the femoral head|
|4A||Indentation type; depression of the superolateral, weight-bearing surface of the femoral head|
|4B||Transchondral type; osteocartilaginous shear fracture of the weight-bearing surface of the femoral head|
|5||Central fracture-dislocation of the hip with fracture of the femoral head|
Although most authors have used Pipkin’s classification since its publication, Brumback and coworkers’ classification is more complete and includes fractures of the femoral head reorganized with associated fractures. Although somewhat cumbersome, its precision warrants the use of this system in future publications.
Another classification of femoral head fractures has been proposed by Müller and colleagues and adopted by the Orthopaedic Trauma Association. Its alphanumeric categories separate and subcategorize split and depression injuries, as well as those associated with femoral neck fracture. For purposes of pooling published literature, this system should also be used when reporting case series or controlled trials.
Femoral Head Fracture and Dislocation
After adequate physical examination, review of the AP pelvic radiograph for location of the femoral head fracture, and evaluation of the femoral neck and acetabulum, early gentle closed reduction is recommended. Closed reduction of the hip is indicated for all hip dislocations regardless of whether an associated femoral head fracture is present. Techniques for closed reduction are outlined in Chapter 52 . Delay must be avoided to minimize the risk of posttraumatic AVN of the femoral head. If a femoral neck fracture is identified, it is probably better to forego any attempt at closed reduction and proceed with open surgery after an urgent preoperative CT scan, if possible. Such an approach may decrease the risk of displacement of the femoral neck fracture with further injury to the vascular supply of the femoral head.
If closed reduction is unsuccessful in the emergency department, then patient should be taken to the operating room for a closed versus open reduction under complete muscle relaxation. A preoperative CT scan, whenever possible, helps alert the surgeon to intraarticular fragments, acetabular or femoral neck fractures, and the size of the femoral head fragment. A delay of more than an hour to obtain a CT scan should be avoided. Percutaneously placing a 5-mm Schanz pin in the proximal femur can provide better leverage than purely manual traction alone for those irreducible dislocations ( Fig. 54-4 ). If unsuccessful, then an open approach should be performed. In general, posterior dislocations should be reduced through a posterior approach. The external rotators and buttonholed capsule are the usual structures blocking reduction. Intraarticular fragments can be removed with this approach, and associated posterior wall acetabular fractures can be operatively reduced under direct vision. Internal fixation of the femoral neck and head, as well as reduction of these fractures, is difficult with this approach. The patient should be placed in the lateral decubitus position to allow access to the anterior aspect of the pelvis should a simultaneous approach be necessary to reduce and internally fix the femoral head fragment. A simple technique of obtaining hip distraction is using a 5-mm Schanz pin placed in the proximal femur. The 5-mm Schanz pin is placed laterally at the level of the vastus ridge. The assistant who will be distracting the hip will stand anterior to the patient. With the hip in neutral extension and the knee flexed at 60 to 90 degrees, the assistant has one arm cradled around the knee and the other hand on the T-handle chuck. The assistant can provide a significant amount of hip joint distraction to allow the surgeon to remove loose bodies and assess the femoral head. This technique allows for more freedom to externally and internally rotate the hip joint to better visualize the hip joint. Another technique is a femoral distractor applied from the iliac crest to the proximal femoral shaft, which helps gain distraction of the hip joint to improve visualization of the reduction. Without performing a surgical hip dislocation, it is difficult to attempt fixation of the femoral head, which is generally medially located. If the surgeon chooses to leave the femoral head fragment unfixed, the patient should be treated by skin or light skeletal traction for 3 to 6 weeks ( Fig. 54-5 ).
Recently Mehta and Routt reported seven cases of irreducible fracture-dislocation of the femoral head without posterior wall acetabular fractures. The physical examination and radiographic findings in this rare injury depicted a different appearance than the standard posterior hip dislocation. The injured limb is shortened, slightly flexed at the knee and hip, and in neutral rotation. Imaging studies showed the suprafoveal femoral head fractures retained in the acetabulum while the dislocated proximal femur is locked against the lateral iliac cortical bone of the supraacetabular region. The dislocated component had buttonholed through the posterior–superior labral and capsular tissues and impacted in the acetabulum. In this series, all seven patients underwent ORIF via a Smith-Petersen surgical approach.
Special Considerations for Patients with Polytrauma.
An unreduced hip dislocation is a musculoskeletal emergency because of the consequences of posttraumatic femoral head necrosis, which increases in incidence with increasing duration of the dislocation. An AP pelvic radiograph is part of the initial evaluation of a patient with multiple injuries and will reveal the hip dislocation plus the femoral head fracture. If the patient is going to the operating room for head, abdominal, or chest procedures, closed reduction of the hip dislocation can be expedited by the orthopaedist’s presence following induction of anesthesia. As soon as muscle relaxation has been achieved and the airway secured, closed reduction of the hip is performed as described in Chapter 52 . If unsuccessful, open reduction should be performed as soon as other lifesaving procedures are completed. If closed reduction is successful, the same algorithm of postreduction CT followed by ORIF of a poorly reduced fracture, débridement of loose bodies, or ORIF of the femoral neck or acetabulum is used. In the case of an associated unrecognized femoral neck fracture or loose bodies, an open procedure should follow as soon as the patient can tolerate a second anesthetic. This open procedure is performed to avoid damage to the articular surfaces in the case of small loose fragments of bone or cartilage and to lower the risk of AVN of the femoral head in the case of the femoral neck fracture. Skeletal traction should be initiated in the interim when loose fragments are identified to minimize damage to the articular cartilage. Delayed operative management has been reported with good functional outcome.
In patients with well-reduced femoral head fractures of the Pipkin type I or II classification, it may be advisable to perform ORIF of the femoral head fragment to allow the patient to be mobilized. Traction, in general, should be avoided in patients with serious thoracic trauma or pulmonary dysfunction. The ability to mobilize patients with multiple injuries has been shown to have the positive benefit of reducing the incidence of pulmonary failure and sepsis.
Indications for Definitive Care.
The goals of treatment for femoral head fractures are to achieve a reduction with residual displacement of 1 mm or less and a stable and congruent hip joint. For isolated Pipkin type I fracture with excellent (less than 1-mm step-off) reduction, closed treatment is recommended. One to 4 weeks of light traction (Buck’s skin traction or skeletal traction) followed by touch-down weight-bearing on crutches for 4 weeks has produced good results in most patients. If the reduction is not adequate, ORIF with small cancellous bioabsorbable or Herbert screws is recommended, using an anterior approach. Herbert screws provide less compressive force across large cancellous surface areas than do standard small-fragment screws. In polytrauma cases, ORIF may also be indicated even when the reduction is good to allow mobilization of younger patients.
The same recommendations apply to type II fractures, but because of involvement of the superior femoral head, only an anatomic reduction on repeated radiographic evaluations should be accepted for conservative care. For cleavage femoral head fractures associated with a femoral neck fracture (Pipkin type III), the prognosis is poor. The prognosis for the injury in regard to posttraumatic AVN of the femoral head is related to the degree of displacement of the femoral neck fracture. For this reason, care must be taken during closed reduction to prevent displacement of a recognized or unrecognized femoral neck fracture ( Fig. 54-6 ). In a younger, more active patient, urgent ORIF of a type I or II femoral head fracture through an anterior Smith-Petersen approach is recommended, along with screw fixation of the femoral neck fracture. Surgical dislocation has been recently reported to provide excellent visualization of the fracture for reduction and fixation. The decision to proceed in this manner should be weighted toward treating those who are active, are physiologically young, and have minimally displaced or nondisplaced femoral neck fractures. In patients who do not fulfill these criteria, a bipolar endoprosthesis or total hip arthroplasty (THA) should be performed.
Pipkin type IV fractures must be treated in tandem with the associated acetabular fractures. The acetabular fracture should dictate the surgical approach, and the femoral head fracture, even if it is nondisplaced, should be internally fixed to allow early motion of the hip joint. Management of the associated acetabular fracture is covered in Chapter 41 .
Femoral head fractures associated with anterior hip dislocations are very difficult to manage. Elevation of the indentation fragment has been advocated, but the long-term results of this technique have not been published. The prognosis is poor because of the risk of posttraumatic arthritis, and the patient should be so informed. Cleavage fractures, if they are large and noncomminuted, may be internally fixed. The repair should be performed from an anterior approach if the CT scan indicates that the major portion of the fragment is anterior and from a posterior approach if the fragment involves the posterior, weight-bearing portion of the femoral head. No results with this treatment have been published.
If closed reduction is successful, a postreduction CT scan is indicated. The scan is then reviewed for reduction of the fragment, status of the femoral neck and acetabulum, and presence of loose bodies. Treatment recommendations are then based on the classification, reduction of the fracture, and general considerations. Traditionally, skeletal traction is applied and 6 weeks of bed rest is recommended. However, complications such as decubitus ulcer, DVT, and pneumonia can occur with prolonged bed rest. Ideally patients that are treated nonoperatively should be allowed to mobilize with touch-toe weight-bearing (TTWB) and regular repeated radiographs should be obtained to evaluate for maintenance of reduction and hip joint congruency.
In combination with closed or open reduction, the indications for fragment excision are severe comminution and interposition of a small femoral head fragment between the femoral head and acetabulum. Excision of the fragment can be accomplished through the same surgical approach used for open reduction. Fragment excision can also be accomplished arthroscopically in the hands of surgeons experienced with hip arthroscopic techniques. If done subsequent to reviewing the CT scan after closed reduction, the open surgical approach is dictated by the location of the fragments. Anterior and inferior fragments should be approached through the Smith-Petersen interval. In the case of interposed fragments, excision is urgent and the procedure must be done quickly to avoid further damage to the articular surfaces.
Open Reduction and Internal Fixation.
ORIF is indicated for all fractures with residual displacement of 1 mm or more, for fractures associated with femoral neck or acetabular fractures, and for fractures with large femoral head fragments that require open reduction of the associated hip dislocation. For most Pipkin type I and II fractures, ORIF should be performed through an anterior Smith-Petersen approach ( Fig. 54-7 ). The surgery is performed with the patient in the “semilateral” position and a large pad underneath the affected hip. These procedures should be performed within several days after closed reduction and a postreduction CT scan. In the case of a posterior approach, fragments off the anterior aspect of the femoral head are difficult to visualize, are harder to reduce, and can be nearly impossible to fix internally. This approach was recommended by Epstein and coworkers because of fear of damage to the blood supply to the femoral head from the anterior capsule. The blood supply to the femoral head from this source is negligible, and because of these operative difficulties, the anterior approach is favored. The patient may be treated in a continuous passive-motion machine postoperatively, along with 8 weeks of touch-down weight-bearing and avoidance of extreme hip flexion (>70 degrees) for 4 to 6 weeks. The anterior approach may be accompanied by heterotopic ossification of functional significance. Such ossification can be avoided by minimizing stripping of the tensor fasciae latae and abductor musculature from the ilium. Indomethacin (Indocin), 25 mg orally three times a day for 6 weeks, or low-dose irradiation may also have a favorable influence, but diphosphonates are probably of limited therapeutic value ( Fig. 54-8 ).
Prosthetic replacement is indicated in a Pipkin type III fracture when the patient is physiologically elderly or the femoral neck fracture is markedly displaced in patients older than 50 or 60 years. Primary femoral head replacement is otherwise contraindicated and should be performed only after a trial of conservative care when the end result of internal fixation is joint incongruity or degenerative arthritis. Should such problems develop, total hip replacement is indicated. Details regarding the procedure of endoprosthetic replacement are discussed under the “Femoral Neck Fractures” section ( Fig. 54-9 ).
Open Reduction and Internal Fixation of Associated Acetabular Fractures.
ORIF of the acetabular fracture in a Pipkin type IV injury is indicated when the fracture is displaced or the hip reduction is unstable. The femoral head fragment should also be internally fixed to gain the benefits of early, relatively unrestricted joint motion. Whereas the main component of the acetabular fracture dictates the surgical approach, the femoral head fracture may require a separate anterior approach to accomplish the reduction and fixation. The details of operative management of an acetabular fracture are described in Chapter 41 .
For most Pipkin I and II femoral head fractures, the preferred surgical exposure is the anterior approach to the hip, Smith-Petersen. This approach allows better visualization and the opportunity to internally fix the femoral head fragment that is often located anterior and medially. The approach does not further compromise the blood supply to the femoral head and thus has no increased incidence of AVN However, there is an increased incidence of heterotopic ossification if there is too much stripping of the gluteal muscle off the ilium.
The intermuscular plane for this approach is between the sartorius and tensor fasciae latae. These structures are innervated by the femoral nerve and superior gluteal nerve, respectively. The deep surgical dissection is between rectus femoris (femoral nerve) and the gluteus medius (superior gluteal nerve). Retracting these structures will expose the hip joint capsule and allow for a T-capsulotomy to be performed.
The patient is positioned supine on a radiolucent operating table with a bump underneath the involved extremity buttock. The C-arm should come in opposite the operative leg. The ipsilateral hip and lower leg should be prepped and draped in a free-leg style using sterile leg stockinette. This will allow for more freedom to rotate (externally and internally) the leg and pulling traction on the hip joint for better visualization.
A longitudinal incision is made starting at the anterior superior iliac spine and extends about 10 cm distally and projecting just lateral to the patella. Identify and protect the lateral femoral cutaneous nerve that exits through the fascia between the sartorius and the tensor fasciae latae. The superficial dissection is made by incising the intermuscular plane between the sartorius and the tensor fasciae latae. The deep dissection is made by retracting the sartorius medially and tensor fasciae latae laterally to expose the direct head of the rectus femoris medially and the gluteus medius laterally. Carefully identify and ligate the ascending branch of the lateral femoral circumflex artery. Exposure of the anterior hip capsule is done by partially releasing the direct (off the anterior superior iliac spine) and reflected head (from the superior lip of the acetabulum), retracting the rectus medially, and retracting the gluteus medius laterally. A T-capsulotomy is performed to provide exposure to the anteromedial femoral head fracture fragment, as well as the femoral neck fracture. With the leg draped, free distraction and rotation of the hip joint is possible manually by pulling on the leg or using a 5-mm Schanz pin in the proximal femur. Closure of the T-capsulotomy is done loosely with No. 2 nonabsorbable sutures after excision or fixation of the femoral fragment.
A standard Kocher-Langenbeck approach can be used in femoral head fractures that are associated with a posterior wall acetabular fracture. This approach is useful in irreducible posterior femoral head dislocations that have a large posterior wall acetabular fracture that requires surgical fixation. The main limitation with this approach is obtaining visualization and fixation of the femoral head if the hip has been reduced in the emergency department. If the hip has not been reduced, then reducing and fixing the femoral head fracture is performed prior to relocating the hip and fixing the posterior wall fragment.
The posterior approach via the Kocher-Langenbeck provides excellent exposure of the posterior wall and posterior column of the acetabulum. There is no true internervous plane with this approach. The sciatic nerve is the main structure that careful attention should be given throughout the procedure.
The patient is placed in a lateral decubitus position on a radiolucent operating table. Using a beanbag for the upper torso will allow imaging and the C-arm should come from anterior to the patient. The ipsilateral hip and lower leg should be prepped and draped in a free-leg style using sterile leg stockinette. This will allow for more freedom to rotate (externally and internally) the leg and pulling traction on the hip joint for better visualization. A second padded Mayo stand should be placed under the operative knee with the hip in neutral extension and the knee in at least 70 degrees of flexion. This will allow less tension on the sciatic nerve.
An incision is made 5 cm lateral to the posterior superior iliac spine (PSIS) and extended obliquely toward the greater trochanteric of the femur. The incision is extended distally the same distance along the lateral femur. The fascia lata and the gluteal fascia are split in line with the incision. Slightly internally rotating the hip will put tension and provide exposure to the piriformis and the short external rotators (superior and inferior gemelli and obturator internus) tendons. Tagging and releasing these structures with nonabsorbable sutures will provide exposure of the posterior hip capsule and the posterior acetabular surface. It is important that the dissection does not violate the quadratus femoris to avoid injuring the medial femoral circumflex artery. Beware of the relationship of the sciatic nerve relative to the piriformis tendon and the short external rotators. Partial or full release of the gluteus maximus tendon insertion on the femur can provide extended retraction and allow easier exposure of the sciatic nerve. This approach provides excellent exposure to the posterior wall, ruptured hip capsule, and labrum to allow easy repair and fixation of these structures. The main limitation will be visualizing and fixing the femoral head fractures that is often located anteromedially and not directly seen if the hip is reduced. In this situation, using a surgical hip dislocation with a trochanteric flip osteotomy, the femoral head can be safely dislocated for exposure and fixation.
Posterior Approach Surgical Hip Dislocation.
Surgical dislocation of the hip with a trochanteric flip osteotomy has been reported to provide excellent exposure for femoral head fractures. In concomitant femoral head and posterior wall fractures, the surgical hip dislocation approach can provide safe exposure for fixation of both fractures via a single approach. This approach provides full visualization of the acetabular socket and allows thorough irrigation and removal of small debris (capsule, labrum, bony fragments) in the hip joint. Careful attention to preserving the deep branch of the medial femoral circumflex artery (MFCA) is the key to minimizing the risk of AVN.
The surgical anatomy is the same as the standard Kocher-Langenbeck approach, but with several important additions. A greater trochanteric osteotomy (digastric or slide osteotomy) is performed to maintain the insertion of the gluteus medius/minimus tendon and origin of the vastus lateralis on the greater trochanter. The vascular anatomy, specifically the course of the MFCA, is important to understand and preserve during this dissection. MFCA can be protected by avoiding the release of the short external rotators because the deep branch of the MFCA courses extracapsularly anterior to these structures before it enters the capsule to provide vascularity to the femoral head. The Z-shape capsulotomy described in the following “Surgical Approach” section is performed to facilitate controlled dislocation of the hip joint and allow full exposure of the femoral head and acetabular socket.
The surgical positioning is the same as the Kocher-Langenbeck approach. Adding a total hip drape will allow placement of the involved leg once the hip is dislocated anteriorly.
A standard Kocher-Langenbeck approach is made. The interval between the piriformis tendon and gluteus minimus is identified. An electrocautery is used to mark a line at the posterior edge of the greater trochanter where the gluteus medius inserts and is extended distally on the posterior border of the vastus ridge. A wafer (1.5 cm) of bone is osteotomized off the greater trochanter in a digastric or slide fashion and the short external rotators are left intact to avoid injuring the MFCA branch. Using a fine-tooth oscillating blade (19.5 × 41 × 0.4 mm) and C-arm imaging can avoid taking too much or too little of the trochanter. Further release of the vastus lateralis distally off the proximal femur (to the level of the gluteus maximus tendon insertion) and remaining fibers of the gluteus minimus will allow the osteotomized segment to be retracted anteriorly with a Hohmann retractor. With the hip capsule exposed, a Z-shaped capsulotomy is made starting at the piriformis and the rim of the posterior acetabulum. The capsulotomy is extended from the posterior acetabulum to the anterior acetabulum and zigzags slightly posteriorly before extending down along the anterolateral femoral neck toward the lesser trochanter. Care should be taken to avoid further damage of the labrum ( Fig. 54-10 ). If there is a posterior wall acetabular fracture, a modification of the capsulotomy should be made to avoid removing the remaining capsular attachment to the wall fragment. The hip is dislocated anteriorly by pulling traction, flexing, and externally rotating the leg and putting the lower leg into the hip holder. Sometimes, carefully using a bone hook around the femoral neck or a 5-mm Schanz pin in the proximal femur may assist with dislocating the hip. The femoral head and the hip socket can be fully visualized for débridement, excision, reduction, and fixation. The femoral head is relocated with manual traction, internal rotation, and extension of the leg. Repair of the Z-shape capsulotomy is performed with nonabsorbable sutures. The greater trochanteric osteotomy is reduced and internally fixed with two 3.5-mm lag screws with washers ( Fig. 54-11 ).
Femoral head fragments that are very small, located caudal to the fovea, and comminuted can be excised. Placement of the 5-mm Schanz pin in the proximal femur can provide better traction than manually pulling on the leg to expose the hip joint. Using a T-handle chuck with the Schanz pin can allow for better grip for traction and rotation of the hip to allow for improved visualization of the fragment or hip joint. Femoral head fragments that are located suprafoveally or large enough to accept screw fixation should be reduced and fixed. Using a combination of small Kirschner wires, small point-to-point reduction clamps, and dental picks, the femoral head fragments can be reduced with minimal further trauma to the articular surface.
Interfragmentary lag screws are often used for definitive fixation of femoral head fracture. This allows for compression of the small fracture fragment and achieves rigid fixation. Multiple surgical implants are available to achieve stable fixation to allow early hip ROM. Commonly used implants are 3.5- or 2.7-mm lag screws from the small and minifragment set, respectively. Ideally, implants are placed outside of the weight-bearing surface, but this is not always possible. Countersinking the screws will be necessary to keep the screw heads flush with the articular cartilage surface. Using the countersink prior to measuring with the depth gauge will allow accurate screw length measurement and allow the screw to be buried below the cartilage surface. Other implant options include headless screws (Herbert, Acutrak, and Synthes headless compression screws), bioabsorbable pins, and cannulated screws. The combination of 3-mm cannulated screws and threaded washers have been shown to do poorly in fixing femoral head fracture. The study reported that the screw backed out into the joint because of the dissociation between the threaded washer and the screw. The average physical component summary in these four patients were worse than the other patients that were treated with other implants.
Treatment of the femoral neck fracture in Pipkin type III fractures takes priority through anatomic reduction of the femoral neck through the Watson-Jones approach and ideally stabilized with three screws (cannulated or noncannulated) in an inverted triangle pattern. The femoral head fracture is assessed after fixation of the femoral neck to determine whether it can be treated through excision, fixation, or nonsurgically. Hemiarthroplasty or THA should be considered as the definitive procedure for elderly patients with Pipkin III fractures. Treatment of the acetabular posterior wall fracture in Pipkin type IV fractures can be performed via the Kocher-Langenbeck approach with or without surgical hip dislocation. The femoral head should be fixed first to obtain a congruent femoral head surface in the hip joint. Using the femoral head as the template, the acetabular posterior wall is anatomically reduced and fixed with a 3.5-mm reconstruction plate (buttress plate). It is not uncommon to have associated labral tears with a Pipkin type IV fracture. If a large labral tear is encountered, a No. 2 Mitek suture anchor along the acetabular rim can be used to repair the tear.
Postoperative Care and Rehabilitation
Follow-up Care and Rehabilitation.
In the situation in which treatment by closed reduction and traction is selected, the 4 to 6 weeks of skin or light skeletal traction should be followed by an additional 4 to 6 weeks of crutch ambulation with touch-down weight-bearing. In general, hip flexion of more than 70 degrees should be avoided for the same period. Ambulatory treatment with these flexion restrictions can be undertaken from the time of hospital discharge with frequent follow-up of the patient to be sure the fragment is remaining reduced despite patient mobilization. At 3 months, supervised active and passive ROM exercises, as well as abduction strengthening, can be initiated.
For ORIF of femoral head fractures, the patient should be immediately mobilized with the flexion precautions noted earlier to limit shear forces and treated with 8 weeks of touch-down weight-bearing and crutch ambulation. Continuous passive motion may be used in the early postoperative period but it has not been confirmed to improve clinical outcomes. Patients should be encouraged to work on passive, active-assisted, and active ROM of the hip to regain motion. Partial weight-bearing should begin at 8 weeks and progress to full weight-bearing thereafter. Gentle strengthening can be started when the patient has regained hip motion. Patients can wean off the crutches completely when they are able to walk without a limp.
In the case of fragment excision, the patient should be asked to limit hip flexion to 60 to 70 degrees for 8 to 12 weeks and should be treated with crutch ambulation during this period, followed by strengthening and motion exercises.
When femoral head fractures are internally fixed in connection with femoral neck or acetabular fractures, early ROM exercises are indicated. The patient should also be treated with touch-down weight-bearing and crutches for 10 to 12 weeks.
Postoperative care in patients who have undergone prosthetic replacement is covered later in this chapter (see “ Femoral Neck Fractures ”).
Chronic instability is most likely to occur in the setting of fragment excision, especially when accompanied by an unreduced or excised acetabular posterior wall fragment. This complication is best avoided by internal fixation of the femoral head and acetabular fragments when they are of adequate size. When instability is recognized early, placement of a posterior wall bone graft with a tricortical iliac crest bone graft can be attempted. Chronic subluxation may result in degenerative arthritis with joint space narrowing, which requires hip arthroplasty or arthrodesis.
Wound infection can result from any operative procedure and, in general, should occur in no more than 1% of patients in whom open reduction of femoral head fragments is performed. Postoperative hip infections are usually occult, so a high index of suspicion is required. Joint aspiration is necessary for early diagnosis. Treatment of deep wound infection must be prompt and includes thorough surgical débridement of necrotic tissue and systemic administration of appropriate antibiotics (see Chapter 22 ).
Heterotopic ossification may follow use of either the anterior or the posterior approach for reduction and internal fixation of femoral head fractures. In Pipkin type IV fractures in which extended surgical exposure is required to reduce and internally fix the acetabular fracture, the incidence of heterotopic ossification may be significant and is related to the approach used (see Chapter 41 ). For Pipkin type I and II fractures, the incidence of functionally significant heterotopic ossification is higher with anterior approaches. Resection of the heterotopic mass 12 to 24 months after injury, when alkaline phosphatase levels are declining toward normal and bone scan activity is decreasing, was the traditional recommendation; however, recent experience indicates this may not be necessary. Patients with posttraumatic heterotopic bone may be treated with resection and ROM therapy when they are medically stable and the bone is mature radiographically with the caveat that the local area should not have active erythema, warmth, or swelling. Although diphosphonates play no role in prophylaxis against this complication, indomethacin, 25 mg orally three times a day, or low-dose radiation prophylaxis may be helpful. Irradiation should probably be avoided in young patients until some long-term follow-up data regarding its use are available.
Sciatic Nerve Palsy
Sciatic nerve palsy occurs in 10% or more of posterior hip dislocations and may thus be associated with femoral head fractures. It may be more common when reduction is delayed, thus adding another reason for prompt reduction. Not infrequently, patients with sciatic nerve palsy have significant dysesthesia during the early recovery period. Should such symptoms develop, they may be helped with gabapentin, amitriptyline, carbamazepine, pregabalin, or a combination of these drugs. Serial electromyograms can yield prognostic information regarding return of function. Ankle dorsiflexion is generally the last function to return, and therefore a posterior splint or plastic ankle-foot orthosis must be used. A dense sciatic nerve palsy that follows a hip fracture-dislocation generally carries a poor prognosis for complete functional recovery.
The incidence of AVN increases with the length of time that the hip remains unreduced. It is also slightly more frequent when hip dislocation is associated with femoral head fracture, probably indicative of the greater degree of trauma required to fracture the femoral head. Treatment is difficult. If the area of subchondral resorption and subsequent fracture is limited, flexion osteotomy may play a role in avoiding hip arthroplasty or arthrodesis in younger patients.
Degenerative arthritis occurs in the vast majority of cases associated with anterior hip dislocation. Similarly, it occurred in about half of the Pipkin type II, most of the Pipkin type III, and about half of the Pipkin type IV injuries reported. Oransky and associates published a long-term follow-up on 21 patients with an average follow-up of 81 months. Nearly all patients (95%) developed posttraumatic arthritis. Treatment of this complication is weight control, walking aids, and antiinflammatory medications. In physiologically older patients, treatment of severe symptoms is total hip replacement. In younger patients with manual labor professions, hip resurfacing or arthrodesis should be considered. In general, THA should be delayed as long as possible or until functional demands decrease.
Assessment of Results
A standardized system for evaluating end results is necessary to facilitate communication regarding treatment and results. Such a system is especially necessary for femoral head fractures because very few surgeons treat more than four or five of these fractures in a career. The system developed by Brumback and coworkers is the most comprehensive system used in the literature, and it is not overly complex ( Table 54-5 ). Because of the association with hip dislocation, optimal follow-up should be a minimum of 3 to 5 years to rule out posttraumatic osteonecrosis of the femoral head.
|Excellent||Normal hip motion, no pain, no significant radiographic changes|
|Good||Seventy-five percent of normal hip motion, no pain, minimal degenerative changes of the hip joint on radiographic evaluation|
|Fair/poor||Painful hip with moderate or severe restriction of hip motion, moderate or severe radiographic joint incongruity, or degenerative joint disease|
Two significant problems are evident when attempting to analyze the published series of femoral head fractures : inadequate follow-up both in the percentage of patients within the series and in duration and lack of a uniform classification. Since Pipkin’s important article of 1957, most authors, including us, have attempted to use his classification. Brumback’s classification is more expansive and complete but has only recently been applied to a series of published patients. This classification should be used in future publications along with the Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association (AO/OTA) classification.
The majority of published cases of hip dislocation with associated femoral head fracture can be classified by the Pipkin scheme. * Femoral head fractures associated with anterior hip dislocation are not included because they do not fall into Pipkin’s categories. Of the 170 cases in the literature, 37 (22%) are type I, 72 (42%) are type II, 25 (15%) are type III, and 36 (21%) are type IV femoral head fractures. Multiple treatment regimens were used for each classification and are discussed independently. Although Pipkin categorized results as excellent, good, serviceable, and poor, the criteria were not clearly defined. Other authors have used a similar softly defined classification of results, which makes comparison of series difficult. Evaluation of these series is made more complex by the fact that many treating surgeons were involved in all series. Limitation of follow-up is a more serious qualifying factor. Because of these problems, conclusions regarding treatment remain uncertain.
* References .Of the 26 Pipkin type I femoral head fractures reported with adequate follow-up, 18 were treated by closed reduction and traction. Of these, 13 showed excellent or good results, 2 were fair, and 1 was poor; two patients were lost to follow-up. The length of traction varied but was generally 4 to 6 weeks. Eight patients were treated by fragment excision because of a noncongruent reduction, fragmentation, or other intraarticular fragments. Two of these fractures had excellent or good results, three were fair, and two were poor; one patient was lost to follow-up. No patient in any published series has been treated with ORIF. Of the 36 Pipkin type II fracture cases published with follow-up, 13 were treated by closed reduction and traction. Of these, eight had excellent or good results, three were fair, and two were poor. Six were treated by closed reduction and excision of the fragment. Four had excellent or good results, and two were fair, with no poor results. Of 17 fractures treated by ORIF, 10 had excellent or good results, 3 had fair results, and 4 had poor results. The large size of the fragment would seem to lend itself to internal fixation. Surgical sectioning of the ligamentum teres facilitates reduction and has not resulted in an increase in poor results. The loss of uniform contact of the femoral head with the acetabulum (the “shim effect”) that occurs with fragment excision adds motivation to attempt ORIF, especially when the position of the fragment on CT reveals a nonanatomic reduction.
A recent randomized controlled trial (RCT) evaluated conservative versus surgical fragment excision of Pipkin type I fractures. Eight patients were randomized into the conservative treatment group, which consisted of closed reduction alone and patients placed into skeletal traction for 6 weeks. The other eight patients had closed reduction and open excision of the femoral head fragment via the Smith-Petersen approach. The postoperative regimen for the surgical excision was the same as the conservative group. Patients were followed from 25 to 60 months with an average of 39 months. Using the Thompson and Epstein score and the Merle d’Aubigne and Postel score, they concluded that the function outcome of the surgical group performed better than the conservative group ( P = 0.032). Careful analysis and limitations (small series, randomization, postoperative protocol of immobilization) of this article must be considered before routinely treating Pipkin type I with surgical excision.
The segmental femoral head fracture classified as Pipkin type III has been reported in 17 patients with adequate follow-up. Three patients received primary arthroplasty because of the anticipated high risk of complications. Another three underwent closed reduction and traction, all with poor results, and one underwent closed reduction and excision with a poor result. Of the 10 treated by ORIF, 5 had excellent or good results, in 2 the results were fair, and 3 had poor results. In this situation, long-term (minimum of 3 to 5 years) follow-up is necessary because of the anticipated complication of AVN, but this information was not available in a significant number of these cases. Of interest is the fact that 5 of 17 cases were situations in which the femoral neck fracture was thought to be produced by the closed reduction. Although these cases may have been simply displacement of nondisplaced neck fractures, the consequences of displacement are significant and the prereduction radiographs should be reviewed carefully to search for a femoral neck fracture. If the attempt at closed reduction requires significant force, the surgeon should proceed with open reduction to maneuver interposed soft tissue out of the way.
Pipkin’s type IV category introduces the variable of acetabular fracture. Management of the femoral head fracture must be included in the overall decision making for the acetabular fracture. Of the 28 type IV femoral neck fractures with associated acetabular fractures and adequate follow-up, 12 were treated with closed reduction and traction; 6 had excellent or good results, the result in 1 was fair, and in 3 it was poor, with 2 patients lost to follow-up. Eight fractures were treated by closed reduction and excision, with no good results, three fair results, and three poor results; two patients were lost to follow-up. Eight fractures were treated with ORIF, with two excellent or good results, one fair result, and four poor results; two patients had no follow-up. The difficulty in evaluating these results is that details regarding classification of the acetabular fracture are lacking but are of extreme importance in evaluating patient outcomes.
Swiontkowski and associates reported 37 cases of femoral head fracture: 17 fractures were Pipkin type I, 9 were type II, 8 were type IV, and 3 were unclassifiable fractures. All but five patients were treated with ORIF, and one patient with a bilateral type IV fracture died. In evaluating anterior versus posterior approaches for internal fixation of type I and type II fractures (12 in each group), the authors concluded that the anterior approach provided superior visualization and a better opportunity to internally fix the femoral head fragment while offering no increase in the risk of femoral head AVN (two cases of AVN occurred with posterior approaches, none with anterior approaches). The incidence of functionally significant heterotopic ossification in Pipkin type I and II fractures treated with the anterior approach was 2 of 12 versus 0 of 12 posteriorly. The cause of the heterotopic bone is stripping of the gluteal muscles off the outer aspect of the iliac wing; the surgical approach now recommended involves the distal half of the Smith-Petersen approach, with the gluteal origin left intact. These results have recently been confirmed in a subsequent patient cohort.
Henle and associates and Solberg and associates reported the largest series (12 patients each) on the outcome of femoral head fracture using the surgical hip dislocation approach. Patients were followed for 2 to 96 months evaluated with the Thompson and Epstein score and the Merle d’Aubigne and Postel score. Ten patients had good to excellent results. Two patients developed AVN and required THA. Heterotopic ossification occurred in five patients. Only two were severe (Brooker III and IV), but had concomitant traumatic brain injury. Solberg and associates retrospectively reviewed 12 patients with Pipkin type IV fractures that were treated with trochanteric flip osteotomy and surgical hip dislocation. Follow-up ranged from 24 to 71 months and outcome was documented with the Thompson and Epstein score and the Merle d’Aubigne and Postel score. Ten patients had good to excellent results. One patient developed AVN of the femoral head and four developed heterotopic ossification (three Brooker II and one Brooker III). The authors concluded that this single approach provided the surgeon good exposure and fixation for addressing both femoral head and acetabular fractures.
Anterior dislocation of the hip associated with a superior indentation or shear fracture is a more recently reported occurrence and is not included in Pipkin’s classification. It is becoming an increasingly recognized phenomenon with acetabular fractures as well. The association of superior femoral head fracture with anterior hip dislocation was initially reported by Funsten and colleagues and subsequently delineated by DeLee and coworkers. The indentation of the superior weight-bearing portion of the femoral head occurs as it levers off the anterior wall of the acetabulum or possibly as it impacts against the superior margin of the obturator ring. Similarly, shear fractures occur as the superior aspect of the femoral head impacts the anterior acetabular rim and is cleaved off. Of the 10 published cases of impaction-type femoral head fractures associated with anterior hip dislocation, 7 had evidence of significant posttraumatic arthritis at follow-up. Of the four cleavage, or shear, fractures, all had significant joint space narrowing at follow-up. It is fortunate that anterior hip dislocations with their associated femoral head fractures are rare because such patients have a high risk of posttraumatic arthritis.
Meta-Analyses and Systematic Reviews
Giannoudis and associates published a systematic review on management, complications, and clinical results of femoral head fractures. There were 453 femoral head fractures with a mean follow up of 55.6 months. The majority of the femoral head fracture-dislocation causes were high-energy mechanisms. Urgent or emergent closed reduction was performed in 81.6 (275/337) of the cases. Successful reduction was accomplished in 84.3% (232-275) of the attempted closed reduction. The other cases had an emergent open reduction because of failed attempts at closed reduction or surgeon’s preference to proceed with open reduction immediately. Definitive management was nonoperative in 22.9% of the cases. Common criteria for nonoperative treatment were anatomic reduction of the fracture-dislocation, no intraarticular fragment, and no instability of the hip joint. When these criteria were not met, operative treatment included excision of the fragment, ORIF, and hip replacement. The authors reported that excision of the femoral head fracture in Pipkin type I fractures had a trend for better results than ORIF, but it was not statistically significant (likely type II statistical error). For Pipkin II fractures, ORIF was recommended, with anatomic reduction and stable fixation being paramount for better outcome. There were only 26 cases of Pipkin type III fractures in this review. The recommendation was ORIF for young patients and arthroplasty for elderly patients. The results for Pipkin type IV fractures are poor and there was no consensus surgical exposure approach to these fractures. The fracture pattern and the overall injury severity characteristics play a crucial role in determining the operative approach. ORIF of the acetabulum to restore the joint congruency should be performed before ORIF of the femoral head. The surgical hip dislocation with trochanteric flip appears to be the best option to address both fractures in Pipkin type IV fractures. The complications obtained from this systematic review were AVN (11.9%), posttraumatic arthritis (20%), and heterotopic ossification (16.8%).
Guo and associates performed a meta-analysis on 10 studies reviewing the complications of femoral head fractures and the surgical approaches. AVN rate was highest in the posterior approach (16.9%), followed by trochanteric-flip (12.5%), and the anterior approach (7.9%). However, the anterior approach had a trend toward higher incidence of heterotopic ossification (42.1%) than the other approaches. They also suggested that the trochanteric-flip appears to be the better operative approach for femoral head fractures.
Femoral head fractures are rare injuries that often occur with posterior hip dislocation. Prompt reduction of the hip is recommended to limit the risk of femoral head osteonecrosis. Definitive management of the femoral head fracture after successful closed reduction of the hip depends on the location of the fracture, the amount of displacement, joint congruity, presence or absence of loose bodies, and associated injuries. Anatomic reduction of the articular surface, rigid fixation, congruent hip joint, and early ROM of the hip is important in limiting the complications of AVN and posttraumatic hip arthritis.
Femoral Neck Fractures
Femoral neck fractures comprise a significant portion of hip fractures. The majority are caused by low-energy trauma in the osteoporotic elderly. However, younger adults exposed to high-energy forces can sustain femoral neck fractures (2% to 5% of the total population of patients with femoral neck fractures). These injuries are a source of great concern because of their potential for AVN of the femoral head. Unfortunately, treatments for this disabling problem are not well suited for young adults. The remainder of this chapter will address intracapsular fractures of the femoral neck. Many of their important features relate to the details of local anatomy, particularly the vascularity and bone structure of the proximal femur. Other factors, such as physiologic age, bone quality, activity level, fracture pattern, and medical comorbidity, play an important role in deciding whether to perform ORIF versus arthroplasty. Recent evidence-based outcome studies and systematic reviews have provided us with a better understanding in how to answer this clinical question.
Once the upper femoral epiphysis is closed, at approximately age 16 years, the proximal femur reaches its adult status. Adults’ neck shaft angle varies little between the sexes and is approximately 130 plus or minus 7 degrees. The femoral neck is normally anteverted with respect to the femoral shaft and has been measured at 10.4 plus or minus 6.7 degrees in normal specimens, again with no difference between the sexes. This, too, is constant after skeletal maturity. The femoral head diameter varies according to the size of the individual, ranging from 40 to 60 mm.
Hoaglund and Low measured the femoral head articular cartilage, which averages 4 mm at the superior portion, tapering to 3 mm at the periphery. A substantial synovial membrane covers the entire anterior of the femoral neck, but only the most proximal half posteriorly. The femoral neck has wide variability in length and shape. The greater trochanter has a large posterior overhang. Thus, the femoral neck lies in the anterior half of the proximal end of the femur when viewed from the lateral orientation. This must be recognized for accurate placement of internal fixation devices. The calcar femorale is a condensed, vertically oriented plate of bone within the proximal part of the femur. It originates in the posteromedial portion of the femoral shaft, radiates superiorly toward the greater trochanter, and fuses with the cortex at the posterior of the femoral neck. This structure has been frequently misunderstood and mislabeled in the hip arthroplasty literature. As pointed out by Harty and later by Griffin, it plays a central role in the development of upper femoral fracture patterns.
The bone density in the upper part of the femur declines with age. Certainly, chronic disease, surgical or biologic menopause, tobacco abuse, and certain medications (i.e., corticosteroids, barbiturates, calcium- or magnesium-binding agents, seizure control medications, and hormonal therapy) adversely affect bone metabolism and presumably the mechanical properties of the proximal end of the femur. Freeman and associates identified trabecular fatigue fractures in the femoral head and at the head-neck junction in cadavers and in specimens removed at surgery (arthroplasty) for femoral neck fracture. Only 1 necropsy specimen (from a 20-year-old patient) did not have any recognizable trabecular fatigue fracture in the upper part of the femur, whereas all 10 of the surgically removed specimens had them. The highest concentration of fatigue fractures (56%) was at the head-neck junction. The threshold density value below which fatigue fractures were observed was 0.5 g/mL. Femoral neck fractures have been similarly associated with declining bone density by Singh and associates and by Sugimoto and colleagues. More than just simple aging seems to be responsible for this.
Vascular Anatomy and Physiology
Trueta and Harrison expanded on the work of Howe and coworkers and used injection techniques to study the vascular anatomy of the proximal end of the femur. The lateral epiphyseal artery, which is the terminal branch of the MFCA of the profunda femoris circulation, supplies most of the femoral head (see Fig. 54-1 ). In 15 of Trueta and Harrison’s high-quality injection studies (barium suspensions examined in 15-µm-thick sections by light microscopy), the lateral epiphyseal artery supplied four-fifths of the femoral head in seven cases, two-thirds in another seven, and slightly more than half in one case. The inferior metaphyseal artery is the terminal branch of the ascending portion of the lateral femoral circumflex artery, and it pierces the midportion of the anterior hip capsule. This vessel supplies the more distal metaphyseal bone anteriorly and inferiorly in two-thirds of the cases studied. The third major blood supply of the femoral head is the medial epiphyseal artery of the ligamentum teres from the obturator arterial system. This vessel generally connects with the lateral epiphyseal artery system. This anastomotic system formed by the two minor vessels may play a role in revascularization of the femoral head after femoral neck fracture. The femur’s nutrient arterial system does not significantly contribute to the proximal part of the femoral neck or the femoral head. The distribution of the minor arterioles from the lateral epiphyseal artery system is preferentially toward the subchondral bone of the femoral head articular surface. Many authors have noted that the important vessels supplying most of the femoral head (the lateral epiphyseal system) are contained within the retinacular reflection at the superior aspect of the femoral neck (the retinacular arteries of Weitbrecht).
Effect of Femoral Neck Fracture on Vascular Supply
A femoral neck fracture produces a devastating effect on the blood supply to the femoral head. Displacement generally correlates with the severity of damage to the all-important lateral epiphyseal artery system. In Sevitt’s series of 25 patients who died after femoral neck fracture, only four femoral heads had a normal vascular pattern. Several authors have noted that after a femoral neck fracture, which compromises the retinacular vessels, the ligamentum teres system provides a source of blood for revascularization of the femoral head by creeping substitution. Focal mechanical failure of the femoral head during this process accounts for the development of segmental collapse in AVN.
Catto examined 188 femoral heads removed at necropsy or at surgery for femoral neck fracture and compared them with 50 control femoral heads. This histologic analysis confirmed that the control specimens had no evidence of marrow cellular changes or osteocyte death. In all 109 femoral heads removed more than 16 days after femoral neck fracture, some damage to the vascular supply was revealed by histologic changes. The cellular changes are detectable from 48 hours on, but it is generally agreed that osteocyte loss proceeds slowly after ischemia and that the cellular changes become irreversible after 12 hours. The slow progression of cellular death was confirmed by incomplete osteocyte “dropout” proximal to the fracture until the third week. Dynamic blood flow studies in adult miniature swine demonstrated that a femoral neck fracture displaced 5 to 7 mm with an osteotome and then reduced anatomically produces a 60% decrease in blood flow to the femoral head.
Although the adverse effect of femoral neck fracture on femoral head blood flow has been documented with certainty, some elements of the situation remain under the surgeon’s control. Optimal reduction of the femoral neck fracture has been shown in numerous studies to be associated with a lower incidence of femoral head AVN. This decreased incidence may be because not all of the vessels of the lateral epiphyseal artery system are torn and reduction either may “unkink” some vessels or, when performed beyond the acute phase, may allow for rapid arterial recanalization. Claffey has shown that a complete, displaced femoral neck fracture can occur without disruption of this critical vascular supply. Gill and associates have demonstrated that if intraoperative drilling of the femoral head produces visible bleeding, AVN of the femoral head will not result. This test during open reduction permits assessment of femoral head perfusion. Additionally, stable internal fixation allows revascularization to proceed in an optimal mechanical environment. Although further vascular damage to the femoral head is unlikely with standard techniques of fixation, Brodetti has demonstrated that the posterior and superior femoral head quadrant should be avoided.
Marked displacement of a femoral neck fracture may disrupt the posterior hip capsule. This is particularly true for high-energy fractures in young patients with normal bone density. When displacement is less than half the diameter of the neck, the hip capsule may remain intact. In this situation, intracapsular hematoma may develop a high enough pressure to occlude the venous drainage system within the capsule, or to limit retinacular arteriolar flow to the femoral head. Several authors have shown that increased intracapsular pressure reduces femoral head blood flow and may cause cellular death. Increased intracapsular pressure does occur after femoral neck fracture. Reduction of femoral head blood flow in association with elevated intracapsular pressure has been confirmed clinically with technetium bone scanning by Stromqvist and coworkers and by intraoperative laser Doppler flowmetry by Beck and associates Most authors have confirmed that extension and internal rotation of the hip elevate intracapsular pressure to a significant degree by limiting capsular volume. This position should be avoided before surgery. The more comfortable position of flexion and external rotation should be encouraged. Given the forgoing findings, hip capsulotomy may improve femoral head perfusion, at least for some patients. Aspiration of the hematoma has been shown to lower intraosseous pressure within the femoral head after femoral neck fracture (an indirect assessment of venous drainage of the head); however, a hematoma will reaccumulate rapidly, and aspiration must be repeated. Ultrasound-guided aspiration after confirmation of intracapsular hematoma has been recommended. This is at best a temporizing measure until capsulotomy and fixation can be definitively provided. Parenthetically, fixation with historical triflanged nails increases intracapsular pressure, and should be avoided.
Hip fracture is an increasing public health problem. It has been the subject of major consensus conferences to set goals for improving care in the last decade and a half. Femoral neck fracture primarily affects individuals older than 50 years of age. Patients younger than 50 years of age account for approximately 2% to 5% of all femoral neck fractures. The incidence of hip fracture in patients younger than 50 years was 3% of the total 3147 patients with hip fractures treated over a 5-year interval in Edinburgh, Scotland, between 1987 and 1991. Femoral neck fracture frequency in younger, active adults involved in vehicular trauma may be increasing in the United States, particularly in association with femoral shaft fractures. It has also been suggested that smaller automobiles with lower dashboards increase the risk of forces being applied to the distal end of the femur in a way that causes fracture of the femoral neck.
The rates of cervical, trochanteric, and subtrochanteric fracture and the overall rate of fracture at all three levels increase with age, are greater for women than men, and are higher in the southern part of the United States. Femoral neck fractures are more frequent in females. Zetterberg and colleagues found the female-to-male ratio for femoral neck fracture over the 43-year period from 1940 through 1983 to be 3.4 to 1. The incidence of femoral neck fracture is higher than can be explained by aging alone. Aging does explain some of the increase, because the mean age of patients sustaining femoral neck fractures has increased from 71.7 to 74.3 years for males and from 72.6 to 79 years for females from 1965 to 1981. The annual incidence of femoral neck fracture for 1000 persons in 1981 was 7.4 for females and 3.6 for males. The increase in incidence is greater in urban (6%) than in rural (3%) populations, which has been confirmed in Great Britain, Korea, and Italy. Most of the literature published on femoral neck fracture is based on population studies done in Scandinavia. Because osteoporosis is associated with fair skin and northern, female smokers, these studies may not be strictly applicable to North American populations. Rates of femoral neck fracture seem to be higher in whites than in African or Japanese populations, and at least a portion of the increased risk can be explained by upper femoral geometry. Although Melton and coworkers did not identify an increasing incidence of femoral neck fracture in the U.S. population, the incidence of the fracture in the late 1970s (9.2 per 1000 persons per year) is not dissimilar. Left-sided fractures may be more common than right-sided fractures for unclear reasons unrelated to hand dominance. There does not appear to be significant seasonal variation. Patients with a femoral neck fracture are at risk for a second hip fracture. Schroder and colleagues found that 68% of second hip fractures were the same type as the first; the mean interval between fractures was 3.3 years, without male-female differences. The risk for a first fracture was 1.6 per 1000 men per year and 3.6 per 1000 women per year, and for the second fracture it was 15 per 1000 men per year and 22 per 1000 women per year. A previous fragility fracture is a strong predictor for any second fracture, including hip fractures. Spine fractures related to osteoporosis are as important a predictor of a hip fracture as wrist fractures, which tend to be more predictive of a hip fracture in elderly white women.
Mechanism of Injury and Prevention
High-energy femoral neck fractures, associated with vehicular trauma or falls from significant heights, are believed to be caused by axial loading of the thigh (by the dashboard in an automobile) with the hip abducted. Such loading will fracture a femoral neck of normal density. Most femoral neck fractures (in excess of 90%) are due to falls from a standing position. Such “low-energy” injuries rarely fracture a femoral neck of normal density. Whether the fracture precedes the fall or the fall causes the fracture is an interesting question. Sloan and Holloway identified 13 of 54 patients (24%) who complained of increasing groin pain before their leg “gave way.” Freeman and associates found numerous histologically evident fatigue fractures in control specimens, with the highest concentration in the subcapital region. Although fatigue fractures of the femoral neck do occur and fairly frequently become displaced, most authors believe that the trauma of the fall plays a role in creating the fracture in most cases. Because the number of fatigue fractures of trabeculae in the femoral neck increases with decreasing bone density, spontaneous femoral neck fractures occur most often with severe osteoporosis. Previous fractures in the same lower limb have been shown to decrease proximal femoral bone density, increasing the risk of hip fracture with low-energy trauma.
Neuromuscular conditions, exclusive of Parkinson disease, are more frequently associated with intertrochanteric than femoral neck fractures. Falls while rising from a seated position are very common and are related to weakness and poor conditioning. Age-related decline in vestibular, visual, auditory, and somatosensory function contributes to these balance problems. Rashiq and Logan investigated the role of drugs as a cause of femoral neck fracture. They suggested that hypnotic or sedating drugs might impair postural control and result in falls. Although the association between femoral neck fracture and sedative or benzodiazepine use has not been consistent, the latest emerging evidence suggests that it plays a role. Excessive alcohol consumption appears to increase risk of hip fracture. The risk of hip fracture in women may be minimized by maintaining body weight, walking for exercise, performing high-intensity resistive exercises, avoiding long-acting benzodiazepines, minimizing caffeine intake, using appropriate footwear, ceasing smoking, and treating impaired visual function and obesity. Moderate alcohol consumption may have a protective effect. A history of stroke and current use of walking aids are associated with an increased risk of hip fracture as is renal failure both in dialysis and renal transplant recipients. Functional dependence on others is associated with an increased risk of falling. Multinutrient or thyroid supplementation alone does not appear to affect weakness or the risk of hip fracture. Vitamin A in high doses likely increases the risk of fracture. A recent meta-analysis has shown that vitamin D supplementation appears to reduce the risk of falls while a population-based study has demonstrated an increased plasma homocysteine level to be a strong and independent increased risk factor for osteoporotic fractures. Postmenopausal estrogen replacement protects against hip fracture in women younger than 75 years. The bisphosphonates are a proven means of increasing bone density and lowering the risk of hip and other nonvertebral fractures. The same is seen with calcium and vitamin D supplementation and growth hormone therapy, even for those patients in residential care. β-Blockers, alone or in combination with thiazide diuretics, has reduced the risk of fractures. The cholesterol-lowering class of drugs known as the statins have been shown to increase bone mineral density, and Pasco and associates have documented a 60% reduction in fracture risk. Coordinated programs of medication review and adjustment, instruction and education, footwear education, and exercise programs prevent falls in the elderly. Training, strength training, and Tai Chi have all been proven to be effective and economically feasible in fall prevention. Even moderate levels of physical activity, including walking, decrease the risk of falling and hip fracture. Patients with dementia and/or living in long-term care facilities require individualized programs for fall prevention. They may return to independent living with intensive geriatric rehabilitation programs. Hip protective devices have been shown to decrease the incidence of hip fracture in nursing home patients, but do not appear to be effective for those living at home. The cost of care for hip fracture patients is significant and with the general aging of the population in much of the world, efforts to prevent fractures are important.
The incidence, mechanism, and prevention of hip fractures are also discussed in Chapter 53 .
Singh and colleagues classified the severity of osteoporosis using trabecular patterns of the intact proximal femur ( Fig. 54-12 ). The radiographic changes were compared with graded iliac crest biopsy specimens and correlation identified. A Singh grade IV or lower represents some degree of osteoporosis. The three-dimensional relationship of primary and secondary compression and tensile trabeculae has been demonstrated from CT data. The radiographic Singh index is related to bone mineral density as determined by dual-energy x-ray absorptiometry (DXA). Ultrasonography has also been shown to correlate with bone mineral density as determined by DXA. The Singh index had fair reproducibility in the same study, with an intraobserver and interobserver κ-statistic of 0.6. Associations have been identified between osteoporosis as defined by this method and severity of femoral neck fracture displacement. Strong correlations between osteoporosis, as determined by DXA, and the incidence and risk of femoral neck fracture have generally been confirmed. DXA can be affected by hip rotation greater than 10 to 15 degrees. DXA (hip) screening has been shown by one study to decrease hip fracture incidence by 36%. Loss of bone mineral density in the first year of treatment with alendronate or raloxifene should not be a reason to stop therapy as most women recover bone mineral density in the second year of treatment ; this phenomenon has been demonstrated to have, as a partial explanation, the phenomenon of “regression to the mean.”
In one prospective cohort of perimenopausal women, patients in the lowest quartile of bone density had a 2.9 times greater risk of fracture in some region of the skeleton; in another cohort, a 2.7-fold increased risk was observed. Wilton and coworkers, using iliac crest biopsies, found only a 2% incidence of this condition in nearly 1000 patients with a femoral neck fracture. In an age-matched population of acutely ill patients without hip fracture, the incidence was 3.7%. The lack of a strong association was confirmed by Hoikka and colleagues and by Lund and associates. Iliac crest bone morphology and bioactivity may not necessarily reflect the morphology and function within the femoral neck. Femoral neck biopsies have shown greater cancellous bone atrophy and fewer osteoblasts and osteoclasts than have iliac crest biopsy specimens in the same patients. Medial femoral neck histologic specimens in patients with femoral neck fracture have shown haversian canals of greater diameter than in age-matched controls. Occult vitamin D deficiency in serum in patients with femoral neck fracture has been confirmed, and vitamin D concentrations within bone in patients with femoral neck fracture have been documented to be significantly lower. Alcohol has been shown to have a negative impact on bone density, possibly through the vitamin D mechanism, as well as through effects on calcium absorption from the gut. Additionally, smoking and high intake of antioxidant vitamins (E and C), as well as vitamin A, have been demonstrated to decrease bone density and increase the risk of hip fracture.
Bone loss is progressive throughout adult life. It affects the hip significantly after 65 years of age, with ultimate strength and load to failure both decreasing with advancing age. Bone density (determined by quantitative CT and dual-photon absorptiometry) was 15% lower than in matched controls in one series of women with hip fractures. In addition, the rate of bone loss in the femoral neck increases with increasing age. Bone loss may preferentially occur in the femoral neck region in patients who sustain femoral neck fractures.
Trochanteric bone mineral density has been shown to be 13% lower in female and 11% lower in male patients with intertrochanteric fracture than in those with femoral neck fracture. Aitken thought that osteoporosis, as determined by metacarpal morphometry, was not a significant cause of hip fracture, even though it might influence the fracture type. Firooznia and colleagues used CT to investigate spinal bone mineral content in a series of 74 women with vertebral fractures, 83 with hip fractures, and 28 with both. Only 4% of patients had spinal bone mineral content below that of their age-matched peers. Osteoporosis has been shown to involve all skeletal sites in patients who are found to have a single vertebral fracture. Although osteoporosis plays a significant role in the severity of fracture displacement and the ability to obtain stable internal fixation, it seems safe to conclude that by itself, osteoporosis is not sufficient to cause femoral neck fractures. Although the level of physical activity before fracture has been proved to have a role (i.e., greater activity leads to a lower incidence) and may well be related to bone density and the quality of trabecular organization, falls remain the initiating factor for the majority of femoral neck fractures. Impact forces after falls exceed the strength of the proximal part of the femur by 50% in older individuals and are approximately 20% less than the strength of the femur in younger individuals. Any fracture between the age of 20 and 50 years increases a woman’s risk of fracture after the age of 50 years. This is likely related to the disuse osteoporosis.
The progression of bone loss from osteoporosis is treatable by hormone replacement therapy ; by the use of calcium and vitamin D, bisphosphonate, and statin drugs; and by high-intensity strength training. In addition, endurance, resistance, flexibility, and balance platform training have been demonstrated to significantly decrease the risk of falling in the elderly population (aged 60 to 75 years). Because a fragility fracture is the best predictor of a subsequent fracture, it is imperative that patients presenting with such fractures receive appropriate evaluation, including DXA scans, and treatment. At this time, most patients are not receiving these services, especially males. Racial disparities in prevention and treatment of osteoporosis have also been identified.
Black males have been shown to have greater bone density at multiple skeletal sites that is not related to body size, mass, or hip axis length. However, the lower risk of femoral neck fracture in Japanese may be explained by upper femoral geometry. Geometric factors that may increase the risk of femoral neck fracture include thickness of the femoral shaft cortex, thickness of the femoral neck cortex, reduction in the index of tensile trabeculae, and a wider trochanteric region. Femoral neck length has been shown to be increasing over time and may be related to femoral neck bone mineral density and generalized osteoporosis. Females have an age-related loss in the femoral neck cross-sectional moment of inertia, which in males is compensated for by increased femoral neck girth. An assessment of hip axis length can be included in a routine DXA evaluation and has been shown to be related to the risk of hip fracture. In one analysis of risk, decreased bone density increased the risk of hip fracture 2.7-fold, whereas an increase in hip axis length increased the risk twofold. Patients with longer hip axes sustain intertrochanteric hip fractures more frequently than femoral neck fractures.
Patients with osteoarthritis of the hip have a lower rate of age-related decline in proximal femoral bone density. In comparing patients with femoral neck fracture and those with intertrochanteric and subtrochanteric fracture, patients with femoral neck fracture tend to be younger and more mobile, are less likely to use walking aids, and more often live independently. When in-hospital clinical outcomes were compared, these patients had shorter lengths of stay.
Consequences of Injury
In addition to pain, impaired mobility, and the occasionally fatal medical complications of femoral neck fractures, nonunion and AVN are notable consequences of these significant injuries.
Nonunion is rare after nondisplaced or impacted fractures. However, after displaced fractures nonunion occurs in 50% to 60% of patients treated nonoperatively, and in 4% to 33% after internal fixation. Several studies have shown that nonunion is a rare problem in patients with normal bone density and in whom stable fixation is achieved. It is most closely associated with increased age and fracture displacement. Unstable fixation with poor reduction is generally the root cause. Most patients with femoral neck nonunions suffer moderate to severe groin or proximal thigh pain and limp, typically with a Trendelenburg gait. Because of pain and impaired gait, most patients will require a reconstructive procedure.
As discussed earlier, femoral neck fractures, particularly if displaced, may damage the blood supply to the femoral head. The resulting postischemic scenario includes revascularization of the femoral head, with subsequent trabecular thinning and collapse. This has been called posttraumatic osteonecrosis, aseptic necrosis of the femoral head, late segmental collapse, and AVN . It occurs in 10% to 15% of patients with impacted or nondisplaced fractures and in 30% to 35% of patients with displaced fractures. Fracture displacement with damage to the arterial supply, as well as intracapsular tamponade, play a causative role. Some reports indicate that patients with normal bone density are at greater risk for this complication. This observation implies that a greater amount of force is involved in causing the fracture and that the displacement and soft tissue injury are therefore greater. Certainly, the incidence of AVN has been reported to be higher in younger adults with high-energy injuries. AVN in young adults has devastating complications because of the limited treatment options as compared to elderly patients. Older individuals with lower functional demands are less likely to but can have symptoms of groin and proximal thigh pain severe enough to warrant a reconstructive procedure in 35% to 50% of cases. Unlike the young adults, total hip replacement is a good option and has consistent good result for elderly patients with symptomatic osteonecrosis. Most agree that the higher the functional demands, the more likely it is that the patient will require a secondary procedure. Nearly all patients younger than 50 years in whom this complication develops require a reconstructive procedure. Their high functional demands and relatively young age make arthroplasty a difficult decision because of the high rates of complications and should be a last resort. The main goal of treatment for young patients with symptomatic femoral head osteonecrosis is to preserve the hip. Reconstructive options include osteotomy to unload the segmental area of femoral head collapse, femoral head core decompression, free vascularized bone grafting, and hip arthrodesis. Hemi-resurfacing of the femoral head is another option but only partially preserves the hip joint. Prevention is the best method for addressing these difficult problems and complications. Early surgery, anatomic reduction, and stable fixation are under the surgeon’s control and should be stressed to minimize further vascular injury to the femoral head. Although controversial, intracapsular decompression is easy to perform and may be beneficial with little chance it can negatively impact the outcome. Close monitoring with periodic radiographs are performed to look for femoral head osteonecrosis. An encouraging radiographic finding of viable femoral head is osteopenia on the injured side as compared with the normal side.
Pain after a femoral neck fracture is minimized by stable internal fixation. Preoperative traction does not reduce pain and is to be discouraged. Parker and Handoll recently reported a review on 10 randomized trials evaluating preoperative traction (no traction, skin or skeletal traction) in hip fractures. They concluded that there is no evidence to support the benefit of traction for pain relief, ease of fracture reduction, or the quality of reduction. Groin or buttock pain that develops in the recovery period is generally associated with loss of stability and impending nonunion, or AVN. However, with modern fixation techniques, and accurate reduction, the incidence of nonunion is well below 10%, and thus the latter is more likely. The pain is probably caused by revascularization of the femoral head with resorption of dead trabeculae and associated microfractures ultimately leading to subchondral segmental collapse. An acute increase in pain without a traumatic event is typical of the final collapse of the segment. Pain can rarely be associated with postsurgical sepsis or injury to the sciatic nerve. In the late stages, pain can be caused by the development of posttraumatic degenerative arthritis, frequently related to segmental collapse and resultant loss of femoral head sphericity.
Limited hip motion is commonly associated with pain. The reason for this is that the position of maximal hip extension is avoided because this position decreases capsular volume and raises intraarticular pressure while placing maximal stress across the femoral neck. This symptom is therefore generally associated with nonunion or AVN. In the remote phase, true loss of motion caused by capsular fibrosis and osteophyte formation is a result of posttraumatic degenerative hip arthritis.
After hip fractures, half or more of patients older than the age of 50 years fail to regain their preoperative level of mobility. In some patients, it may result from complications of the fracture; in others, it may be from deterioration of their overall mental or physical condition. In many cases, this results in loss of independent living for an older patient. Successful return to independent living after hip fracture is best predicted by preinjury function in activities of daily living, absence of medical conditions that limit rehabilitation, and cognitive function. Women have lower mortality after hip fracture than do men from the first year throughout the first decade after hip fracture.
It may be assumed that differences among societies and medical care systems will influence the course and site of rehabilitation after femoral neck fracture. No convincing data have been published that demonstrate the superiority of any given form of treatment of femoral neck fracture with regard to outcome measured by the ability to walk or by the rate of institutionalization. Rehabilitation strategies to improve hip fracture outcomes have been identified as a research priority. Patients with senile dementia have significantly worse functional outcomes and higher mortality rates.
Medical complications associated with femoral neck fractures increase with age and severity of medical comorbidities of the patient at injury. One prospective review demonstrated that major medical complications occurred in 9% of patients who were healthy before the fracture and in 21% of patients with medical comorbidities (64% of these patients died). Potential complications include urinary tract infection, wound infection, ileus (occasionally with a risk of cecal rupture), mental status changes, stroke, myocardial infarction, pneumonia, DVT, pulmonary embolism, and death.
For older patients with hip fractures, a full medical evaluation must be performed and treatment instituted to deal with dehydration, electrolyte imbalance, and pulmonary dysfunction. Delirium is common in patients with hip fracture. One study revealed a prevalence of 9.5% early in the hospitalization and 53% postoperatively. The risk of medical complications is favorably influenced by early surgery and mobilization. However, the findings of Kenzora and coworkers and Eiskjaer and Ostgard of a higher rate of mortality in patients undergoing surgery the first day after injury emphasize the need for adequate medical evaluation and preoperative treatment of correctable medical conditions. Treatment protocols have been developed that use physiologic status scores to select patients with better physiologic status for internal fixation.
DVT is not uncommon after hip fracture, with one series reporting a rate of 23% in patients treated with standard means of prophylaxis (aspirin or warfarin). The vast majority of investigators have identified prophylaxis as a favorable influence on the rate of DVT. Dextran, warfarin, subcutaneous heparin therapy, phenindione, aspirin, dihydroergotamine, low-molecular-weight heparin, and intermittent compression boots have all been reported to decrease the incidence of DVT. Limb elevation and early patient mobilization also favorably influence the rate of thrombosis. Some form of prophylaxis against DVT and pulmonary embolism should be instituted in the preoperative or early postoperative period. Ultrasonographic scanning is the diagnostic technique of choice for the diagnosis of lower extremity DVT because of its high sensitivity and specificity. Despite agreement on the appropriateness of DVT prophylaxis, significant variation in use remains apparent.
A low serum albumin level and total lymphocyte count is associated with a higher risk of mortality and poor functional outcome in older patients. Nutritional supplements have been shown to play an important role in aiding recovery and minimizing wound healing complications. Medical consultation should be sought preoperatively and again with any sign of postoperative complications to minimize the effect of these problems (see Chapter 53 ).
An increased mortality rate over that of the general population after femoral neck fracture has been confirmed in numerous studies. Higher rates of mortality are apparent for patients with medical comorbidities and for males. There is some evidence to suggest that smaller rural hospitals have higher in-hospital mortality rates than larger urban hospitals. In the large series of Barnes, the mortality rate in the first month after surgery was 13.3% in men and 7.4% in women. The mortality rate increased significantly when the surgery was delayed beyond 72 hours. Similarly, in a large Norwegian series published by Dahl, the figures were 17.1% for males and 9.8% for females in the first month after fracture. When compared with an age-matched population, the mortality rate was 15 times greater in the first month and 7 times greater in the second month and thereafter followed the population trends. Kenzora and colleagues found a mortality rate of 13% at 1 year in the femoral neck fracture population versus 9% for age-matched controls. Eiskjaer and Ostgard identified the following factors (in order of decreasing importance) as influencing mortality in a series of 204 patients treated with cemented bipolar hemiarthroplasty: cardiac factors, status as a nursing home patient, chronic pulmonary disease, serum creatinine value greater than 1.7 mg/dL, pneumonia, previous myocardial infarction, duration of surgery, and gender. These medical comorbidities do not completely explain the increase in hospital mortality risk in patients with femoral neck fracture. Low admission hemoglobin (<12.0 g/dL in females, <13.0 g/dL in males) is a marker of chronic disease and has been identified as a significant predictor of higher length of hospital stay and increased 6- and 12-month mortality. Diabetes increases the risk of in-hospital mortality. Darner has developed predictive mortality scales and has determined that impaired mobility after fracture is the highest or best predictor of death. The finding of higher mortality in nursing home patients has been confirmed in the North Sydney “fractured neck of femur” outcomes project. The following factors had no influence on mortality: age, time delay from admission to surgery, mode of anesthesia, and cerebrovascular disease. Equivalent mortality rates for patients who had a prefracture cerebrovascular event has been confirmed by Youm and associates. The overall mortality rate in their series was 20% at 6 months and 28% at 1 year. Obesity may lessen the risk of dying. Holmberg and coworkers confirmed the clinical suspicion that patients who sustain their femoral neck fractures in institutions have higher mortality (three times) than do those who are injured at home. Similarly, patients with senile dementia have a mortality rate 21% higher than controls who do not have dementia. Dementia, delirium, and depression all increase the risk of mortality. Patients with multiple medical complications after surgery have the highest risk of dying.
Patients with a second femoral neck fracture have a higher mortality rate than is seen after a first such fracture. This was confirmed by Boston, who found a 3-month mortality rate of 30% after the second fracture versus 13% for a single such injury. Several studies have suggested that the mortality rate is higher after prosthetic replacement than after internal fixation. In addition, Chan and Hoskinson found a higher mortality rate after a posterior approach for prosthetic replacement (20.6%) than after an anterior approach (6.5%). Despite the high mortality rate after femoral neck fracture, 1-year mortality rates are higher with intertrochanteric fractures than with intracapsular fractures (38% vs. 29%, in one study). Poor nutrition as a factor in mortality impacts men more than women. Smoking and alcohol abuse affect both sexes equally. Patients in their 90s and older do not have higher mortality rates, but are at greater risk for losing independence and function. Deaths after hip fracture are frequently attributable to venous thromboembolisms (VTEs), and nutritional and pulmonary dysfunction. Thus, VTE prophylaxis, supplemental nutrition, and early mobilization of the patient are warranted.
In summary, minimizing medical complication goes beyond just performing surgical fixation or arthroplasty for elderly hip fractures. Multidisciplinary and comanagement approaches, such as fracture liaison services between the orthopaedic surgeon and medical specialist or hospitalist, are imperative to improving care of the geriatric hip fractures. Preoperative considerations should also include the involvement of nursing staff, rehabilitation specialists, nutritionists, social workers, pain management specialists, and family members (if available). Integration of all these crucial healthcare providers can significantly reduce the high morbidity and mortality that often accompany these injuries.
Commonly Associated Injuries
In the “high-energy” femoral neck fracture population, associated injuries are common. Most series involving patients younger than 50 years with nonpathologic femoral neck fractures report an incidence of head, chest, abdominal, or extremity fractures or dislocations in the range of 50% to 60%. Closed head injury, cervical or thoracic spine fractures, pneumothorax or hemopneumothorax, and splenic or bowel injury occur commonly in association with a high-energy femoral neck fracture. Because of the axial loading mechanism, the most frequent musculoskeletal injury associations are ipsilateral tibial or femoral fractures, patellar fracture or knee ligament injury, and ipsilateral pelvic or acetabular fracture or hip dislocation. In the more common “low-energy” femoral neck fractures, related to falls from a standing position, associated injuries are less common. Head injury, including subdural and epidural hematoma, may occur occasionally. Ipsilateral injury to the upper extremity (commonly the distal end of the radius or the proximal part of the humerus), which occurs in an attempt to break the fall, is seen in 1% to 2% of cases. Far more common are medical problems, such as a cerebrovascular accident or myocardial infarction being responsible for the fall.
In the most common setting of low-velocity femoral neck fractures resulting from falls, clinical suspicion is based on the history, the complaint, and the results of physical examination. With high-velocity injuries, the patient may be unconscious or distracted by other sources of pain. Physical findings, such as leg deformity because of a displaced femoral shaft fracture, may also hide a femoral neck fracture. Thus, radiographic assessment of the hip and pelvis is essential for all patients with hip pain, or with potentially associated injuries, as discussed later on in this chapter.
In the case of a young patient involved in a high-velocity crash, information about the event is helpful in directing the clinical and radiographic evaluation. The physical and radiographic examinations, however, are usually more valuable than the history for directing evaluation and treatment. An accurate history may contribute more to the evaluation of the much more frequent low-energy femoral neck fracture of older individuals. Information concerning medications, medical conditions, and activity level is important. One should always ask whether the patient had pain in the groin or proximal part of the thigh before the fracture, as such pain may suggest the presence of some type of pathologic fracture.
For these elderly, occasionally demented patients, it is essential to obtain a reliable description of the patient’s activity level before injury. Did she or he walk independently? Was the patient able to assist with transfers from bed to chair? This information helps the physician choose between internal fixation of the fracture and prosthetic replacement. Overall prognosis is influenced by the patient’s preinjury functional level, preexisting medical problems, and, especially, mental status.
In addition, in an older patient, it is critical to evaluate risks and treatment opportunities for osteoporosis and for fall prevention. Switzer and associates reported suggestions on overcoming barriers to osteoporosis care in order to minimize future fractures. Effective osteoporosis care for the vulnerable hip fracture patient starts with identification, diagnoses, and workup of hip fracture patients in the acute care hospital setting. This can be initiated by an established coordinator-managed program, system modification, or automatic specialist referrals. Ultimately, vulnerable hip fracture patients are discharged on appropriate supplements and medication. Caregivers and primary care physicians should be made aware and appropriate follow-up should be made to ensure or improve adherence with the treatment recommendations.
Evaluation of a patient involved in high-energy trauma is covered in Chapter 9 . For femoral neck fractures associated with minor falls, the examination can be conducted in a more focused manner. The presence of a femoral neck fracture may be apparent from the attitude of the affected leg, such as obvious shortening, external rotation, and reluctance to move the limb. Alternatively, usually with an undisplaced fracture, the injury may be occult and suggested only by the patient’s complaints of groin, thigh, or (rarely) lateral hip pain. Tenderness in the hip region produced by percussion on the sole of the foot with the fist or pain at the extremes of hip motion, particularly rotation, may be the only local physical findings to suggest such an occult hip fracture.
After evaluating vital signs and mental status, the head and neck examination should focus on areas of tenderness, evidence of contusions or abrasions, and decreased cervical ROM. If any of these findings is present, a hard cervical collar should be applied until the cervical spine can be cleared radiographically. The chest should be palpated for signs of rib fracture and auscultated to rule out pneumothorax. A screening examination of the upper and lower extremities should follow, with palpation and with the patient, if alert, putting the joints through a ROM. The examination should then focus on the affected lower extremity. The trochanteric region should be evaluated for contusions and for traumatic or nontraumatic skin conditions that might influence surgical management. The knee should be examined for tenderness, effusion, and instability. If such evaluation is not possible because of thigh pain, the knee examination must be repeated before the surgical procedure but after initiation of anesthesia. The thigh and leg should be palpated and the foot and ankle examined for signs of trauma. The circulatory status of both limbs should be assessed as a baseline for follow-up evaluations and the status of the pulses carefully recorded. Finally, a complete sensory and motor examination of the limb should be performed and the findings on physical examination detailed in the medical record.
Radiographic evaluation should include AP and lateral plain radiographs of the entire femur as well as an AP radiograph of the pelvis. For patients involved in high-energy injuries, an AP radiograph of the pelvis is routinely obtained as part of the early trauma series of radiographs. The proximal femur should always be carefully scrutinized. When possible, the legs should be taped at the ankle level to hold the hips in internal rotation before obtaining this film. Physical or radiographic findings of ipsilateral leg trauma, or the presence of hip pain, is also an indication for a routine AP pelvic radiograph.
In the low-energy setting, with history or examination suggesting a hip fracture, an initial AP pelvic radiograph, with the legs taped or held in gentle internal rotation, should be obtained and carefully inspected for a femoral neck fracture. If a hip fracture is suspected but is not evident on this initial radiograph, proceed with imaging studies as discussed later. Nonpathologic high subcapital fractures can be misinterpreted as pathologic fractures in 17% of plain radiographs. An estimate of osteoporosis should be made with the Singh index (using the intact contralateral hip) because this index has some predictive value regarding the degree of osteoporosis and the potential for obtaining stable internal fixation.
If a fracture is evident or suspected on clinical grounds, a cross-table lateral radiograph of the affected limb is also required. This radiograph is made with the affected limb remaining on the stretcher while the good limb is flexed up and out of the x-ray beam. The lateral view should be scrutinized for posterior femoral neck comminution because this view likewise affects the prognosis for obtaining stable internal fixation. Obtaining a traction and internal rotation radiograph has been shown to improve resident ability to classify proximal femur fractures. Koval and associates reported that if properly performed by a trained physician, this traction view can better delineate the fracture and assist with surgical preoperative planning. The fracture patterns seen in elderly patients are different from those seen in young patients. Elderly patients with poor bone quality and low-energy injury mechanisms will usually sustain a femoral neck fracture that will often be subcapital. Commonly, the fracture pattern is a transverse fracture with impaction at the fracture site. In contrast, young patients with good bone quality and high-energy mechanisms will often sustain a basicervical or more distal neck fracture. The fracture pattern is more biomechanically unstable because of the vertically oriented fracture position. These characteristics have important implications when obtaining and maintaining stable fixation, both of which are necessary for healing to occur.
Search for Occult Fractures.
When the patient’s history or examination suggests a femoral neck fracture but radiograph findings are lacking, special evaluation is necessary. The initial study should be an AP pelvic radiograph, with the legs taped or held in gentle internal rotation. This should be carefully inspected for a femoral neck fracture. If a hip fracture is suggested from the history and physical findings but is not evident on this initial radiograph, it is important to obtain an AP radiograph of the symptomatic hip with the femur sufficiently internally rotated to show a maximal profile of the anteverted femoral neck region because this projection will be the most likely to demonstrate an occult fracture. In the absence of an obvious proximal femur fracture, this mandates further evaluation if the clinical setting is suspicious. High-quality CT scans with appropriately oriented reconstructions typically demonstrate a fracture, if even slight displacement is present. They may have been routinely obtained as part of the initial evaluation of a polytraumatized patient. An MRI, if the patient can tolerate it, has the advantages of demonstrating intraosseous swelling in acute undisplaced fractures as well as providing an immediate answer that expedites patient disposition and treatment. MRI is now the preferred imaging study to definitively evaluate for occult hip fracture. Iwata and associates reported that the T1-weighted coronal MRI scan showed a hip fracture with 100% sensitivity and is the best sequence to diagnose occult hip fracture ( Fig. 54-13 ). In the past, technetium-methylene diphosphonate (MDP) bone scans have been used to diagnose occult femoral neck fractures, but they often do not become positive for 2 to 3 days after injury in elderly patients.
In the case of a high-energy injury, clinical suspicion for occult femoral neck fracture should be raised by an associated patellar fracture or knee ligament (especially posterior cruciate) injury, femoral shaft fracture, or an ipsilateral calcaneus, distal tibia, or tibial or femoral shaft fracture resulting from vehicular trauma or a jump or fall from a significant height. CT scans for screening for abdominal injury are often available and a few cuts of the femoral neck are generally available. If high-quality films of the hip do not reveal a femoral neck fracture and the associated injuries require urgent care, the physician should proceed with the urgent procedure. When treatment of other injuries is not pressing, the just-mentioned studies (i.e., CT or MRI) might be considered preoperatively. See later in the chapter for evaluation and management of ipsilateral femoral neck and femoral shaft fractures.
CT and earlier tomographic techniques are helpful in identifying a nondisplaced femoral neck fracture when it is not radiographically apparent but is suspected on clinical grounds (e.g., history, complaint, or pain on rotation of the hip). CT may be particularly useful in a patient with a high-energy fracture who is undergoing an examination for abdominal, pelvic, or spine trauma. This technique is also of use in differentiating pathologic (nonosteoporotic) femoral neck fractures from nonpathologic fractures. Although good studies have confirmed that stability of the postfixation construct is most dependent on bone density, as yet, no clinical data derived from CT-measured density can be used to determine the potential for stability of internally fixed femoral neck fractures.
Radionuclide uptake has been studied for more than 50 years in the hope that such techniques will detect AVN at an early stage. Historically, radioactive calcium, phosphorus, and iodine were injected and Geiger counter systems used to obtain data that could be assessed in terms of femoral head blood flow. Unfortunately, these studies were inconsistent, and involved high radiation burden for the patient.
The two techniques that remain in use for assessment of femoral neck fractures are sulfur colloid scans and technetium 99m ( 99m Tc)-diphosphonate scans. The former is a compound that demonstrates bone marrow viability and has shown some success as a predictor of AVN. Unfortunately, a large series of patients has not been monitored following the use of this technique. It has been considered to be accurate, and the data obtained have been used to decide whether to perform prosthetic replacement. Compared with 99m Tc-diphosphonate scanning, it has the drawbacks of a significantly higher radiation burden, less detailed images, and a longer delay after injection before the images are obtained.
99m Tc-diphosphonate scanning is a method for obtaining predictive data in regard to the risk of nonunion and AVN. Stromqvist and coworkers published results from 468 patients who underwent serial bone scans after femoral neck fractures. They determined that visual evaluation of pictorial images (scintigrams) is neither reproducible nor reliable. However, quantitative scans (scintimetry) allow side-to-side digital comparison using carefully defined regions of interest. Recently, Kim and associates reported the role of preoperative bone scan to determine if the status of blood flow to the femoral head could be evaluated prior to surgery. This information was considered to be potentially useful to decide whether the patient should have an internal fixation or arthroplasty. They concluded that preoperative bone scan was not beneficial in determining treatment method. While preoperative bone scans were not predictive of AVN, such studies done 2 to 3 weeks after injury and fixation can identify patients likely to develop these problems. Data on femoral head uptake are expressed as a ratio relative to the intact side. If the ratio of the injured to the uninjured region of interest is less than 0.90, patients had an 84% chance of nonunion, AVN, or both. For patients with an abnormal contralateral hip, the uptake ratio of the femoral head to the ipsilateral femoral shaft can be measured. If this ratio is at least 0.40, uneventful healing is likely. By 4 weeks after injury, those in whom nonunion or AVN eventually develops have equal or higher femoral head uptake, which lasts up to 24 months after injury. These findings have been duplicated in animal models. These data integrate well with the hypothesis that revascularization follows an avascular episode and produces trabecular resorption and weakening, which allows mechanical failure in the subchondral plate and femoral head collapse.
Scintigraphic images (nonquantitative) produced by nuclear medicine scanning can be helpful for diagnosing occult problems of the proximal end of the femur. When plain radiographs are normal, osteoblastic activity in the region of the femoral neck, shown by a technetium-labeled phosphate scan, may indicate a nondisplaced or fatigue fracture. These changes may not appear until 72 hours after onset of symptoms. Scintigraphy can also help in searching for occult metastatic lesions, which might affect treatment plans for a pathologic fracture. These scintigraphic images are not, however, helpful for preoperatively establishing the prognosis of a given femoral neck fracture ( Fig. 54-14 ).
Dual-photon absorptiometry has been shown to provide an accurate determination of spinal osteoporosis. However, some published reports indicate that spinal bone density is not directly related to femoral neck bone density. The technique has been adapted to more accurately determine femoral neck bone density, and data regarding the bone density necessary to achieve stable internal fixation are available. Incorporation of hip axis length as an additional predictor of hip fracture risk is also feasible.
Magnetic Resonance Imaging.
MRI has been shown to be a sensitive indicator of avascular change within the femoral head in nontraumatic forms of AVN. The ability of MRI to detect acute femoral head vascular changes has been noted to be somewhat limited in clinical as well as in vivo laboratory studies. Severe distortion of the images is produced by proximal femoral internal fixation devices. Pure titanium or nearly pure titanium fixation devices must have been used if one wishes to obtain satisfactory femoral head MRIs after internal fixation; dark shadows remain on the images, but distortion is minimized. This technique may ultimately allow a greater understanding of the pathologic processes of posttraumatic AVN; however, it remains a tool that is unable to predict the patients in whom femoral head collapse will develop from posttraumatic osteonecrosis. It is the method of choice for detecting nondisplaced femoral neck fractures in patients with symptoms and normal radiographs because of the delay (up to 72 hours) in radionuclide uptake in fresh fractures and the suboptimal uptake of technetium-labeled phosphate compounds in the most elderly patients. MRI images can also help define the presence and extent of metastatic tumors involving the femur.
High-resolution, real-time ultrasonography has attracted interest as a simple, repeatable, noninvasive means of diagnosing DVT. A variety of criteria have been proposed for the ultrasonographic diagnosis of venous thrombosis. In a prospective study of 40 patients with hip fractures, Froehlich and coworkers used a single test: noncompressibility of the vein lumen. This test was applied from the calf veins to the common femoral artery and validated by venography. Five patients (12.5%) had major thrombi; all were asymptomatic. Compression ultrasonography had an accuracy of 97%, a sensitivity of 100%, and a specificity of 97%. It is well tolerated and is significantly less expensive than venography, but it requires a well-trained technologist.
Malnutrition is more common than what might be expected in elderly patients with hip fractures, even in those with adequate resources. Improved outcome after hip fracture has been demonstrated when dietary supplements are provided. Screening studies that may be used to evaluate nutritional status are the absolute lymphocyte count, serum albumin, transferrin, and skin fold thickness.
Diagnostic algorithms for patients with multiple injuries are presented in Chapter 9 . To prevent missing a femoral neck fracture (which occurs most frequently in midshaft femoral neck fractures), the femoral neck must be carefully scrutinized, as outlined previously. Spine and extremity radiographs are obtained in accordance with findings from the initial assessment, which includes a history of the injury and physical examination. A preoperative electrocardiogram, chest radiograph, complete blood cell count, urinalysis, and serum electrolyte, creatinine, and serum albumin studies should be obtained in all patients older than 40 years of age (see Chapter 53 ).
The differential diagnosis for a femoral neck fracture in a high-energy trauma patient must include pelvic fracture, acetabular fracture, hip dislocation, intertrochanteric or subtrochanteric femoral fracture, isolated trochanteric fracture, and a contusion or muscle avulsion without fracture. The differential diagnosis of a patient with a low-energy femoral neck fracture should include intertrochanteric or subtrochanteric femoral fracture, pelvic fracture, acetabular fracture, isolated trochanteric fracture, and hip contusion or traumatic trochanteric bursitis. Both the history and the results of physical examination must be used to evaluate the potential for pathologic lesions, nondisplaced fractures, fatigue fracture of the proximal part of the femur or pelvis, and hip arthritis.
Evolution of Classification Systems
Senn’s 1889 recommendation of immediate reduction and fixation for femoral neck fractures created a need for a classification system with which to compare and report results. Speed urged the formation of study groups for this fracture, noting that “in comparison to practically all others this fracture remains unsolved.” Pauwels (1928) classified femoral neck fractures according to the fracture inclination. Pauwels type I (<30 degrees) is a horizontal fracture that, because of minimal shear forces, has the lowest risk of nonunion. Pauwels type II (angle between 30 and 50 degrees) has an intermediate inclination, and the Pauwels type III (>50 degrees) fracture is a more vertical fracture ( Fig. 54-15 ). Because the more horizontal fractures tended to be impacted fractures and the more vertically oriented fractures were generally associated with higher energy and displacement, this classification was somewhat prognostic. Pauwels’ concept was incorporated into the proximal femoral section of the AO comprehensive classification of fractures, also adopted by OTA.
Garden attempted to classify femoral neck fractures according to their prognosis and incidence of complications. His grade I is an incomplete fracture impacted in a valgus position, grade II is a nondisplaced fracture, grade III is a fracture displaced in a varus position, and grade IV is a completely displaced fracture, recognizable by the trabeculae of the free-floating femoral head being realigned with the trabecular pattern of the supraacetabular pelvis ( Fig. 54-16 ). It is probable that many, if not most, grade I injuries are actually complete fractures impacted in a valgus position. Nondisplaced, nonimpacted fractures are only occasionally seen. Because of their high risk of displacement, these grade II fractures deserve their separate category. The difference between Garden grade III and grade IV displaced fractures is frequently difficult to delineate on radiographic review. Furthermore, Barnes’ large series failed to demonstrate a significant difference in the risk of nonunion and AVN between these two groups. Considering results and incidence of complications in the Barnes series, the Garden grades I and II were similar, as were grades III and IV. The significant differences were between displaced and nondisplaced femoral neck fractures. This finding has been confirmed by Parker. The AO/OTA comprehensive classification incorporates these concepts as well as those of Pauwels. It has been proposed that this system can be used to select appropriate methods of management; however, this application has not been validated in a series of patients.
Femoral neck fractures with associated femoral shaft, femoral head, or acetabular fractures need to be classified separately. In the case of ipsilateral femoral shaft/femoral neck fracture, the risk of AVN is 5%, far lower than without the associated shaft fracture. Casey and Chapman suggested that much of the energy producing these combined fractures is dissipated in the femoral shaft, so that the femoral neck injury is a lower-energy fracture. Femoral neck fractures associated with femoral head fractures carry a very poor overall prognosis in terms of their high risk of AVN and joint degeneration. These rare injuries are therefore included in the classification of femoral head fractures. Finally, the outcome of femoral neck fractures associated with acetabular fractures depends more on the pattern of the acetabular fracture. These injuries are therefore classified with acetabular injuries (see Chapter 41 ).
Nondisplaced Femoral Neck Fractures.
Nondisplaced fractures include both truly nondisplaced fractures and those impacted into a valgus position. Their absence of displacement is associated with lower risks of nonunion and AVN. However, the functional impact of these fractures remains significant. Biologically, their better prognosis results from the femoral head’s main arterial supply seldom (if ever) being disrupted with these fracture patterns. It may be, in these low-energy, undisplaced fractures, that intracapsular tamponade plays more of a role in producing AVN when it occurs.
Displaced Femoral Neck Fractures.
Displaced fractures include all femoral neck fractures with any detectable displacement. In the strictest sense, such displacement refers to any alignment offset (translation across the fracture plane) between the distal intertrochanteric fragment and the proximal femoral head fragment. This displacement is important for prognosis, because it correlates with damage to the major arterial supply to the femoral, and thus with risks of both AVN and nonunion. Additionally, when fracture management is delayed, synovial fluid may bathe the fracture surfaces, adding another impediment to fracture healing. The increased risk of AVN and nonunion after displaced femoral neck fractures should be considered when choosing treatment, and favor selection of arthroplasty, especially for older, more sedentary patients.
Fatigue Fractures of the Femoral Neck.
Fatigue fractures result from repetitive loading in pathologic (rheumatoid arthritis, osteoporosis) and nonpathologic (military recruits or repetitive trauma related to running) bone. Devas subclassified stress, or fatigue, fractures into two subgroups, transverse (tension) and compression, based on their appearance, in recognition of their different prognoses. It is important to recognize that fatigue fractures occur more often in patients with osteoporosis than in those with normal bone density. An association between stress fractures of the femoral neck and recent total knee replacement has been repeatedly identified. Stress fractures of the femoral neck are generally associated with decreased bone mineral density, even in younger patients.
Transverse (Tension) Fractures.
Transverse fractures start as a radiolucent crack at the superior-lateral part of the neck. They are best demonstrated on AP radiograph with the hip internally rotated. They typically progress to completely displaced fractures, over days to weeks. If left untreated, patients in this fracture subgroup have a significant risk of displacement. These fractures are distinct from nondisplaced acute fractures in that they are not associated with a single traumatic event. The functional impact of these repetitive-use injuries, even when treated optimally, is significant, particularly in young patients. Displacement of the fracture should be avoided by prompt surgical fixation with multiple screws to minimize the risks of displacement with resulting AVN and osteoarthritis. In one series of displaced femoral neck stress fractures in military recruits, AVN developed in 28%; delayed treatment and varus malalignment were strongly associated.
Compression fractures are seen radiographically as a haze of intraosseous callus in the inferior–medial part of the neck. They have essentially no risk of displacement without additional trauma, so most authors agree that these fractures should be treated by partial weight-bearing, using crutches until the patient’s symptoms have completely resolved.
Pathologic Femoral Neck Fracture.
Treatment of pathologic fractures, including those of the femoral neck, is covered in Chapter 20 .
Femoral Neck Fractures in Young Adults.
Treatment of fractures of the femoral neck in patients with open proximal femoral physes is beyond the scope of this text. (See Skeletal Trauma in Children , ed 5, Chapter 13 .) In adolescents and adults younger than 50 years of age, femoral neck fractures do occur as a result of high-velocity trauma in persons with normal bone density. The same classification (displaced vs. undisplaced) applies as noted previously, but the prognosis is worse in these younger patients. In these individuals, extreme trauma is required to produce displacement of the fracture fragments, which explains the increased incidence of AVN and nonunion in series of young patients. Smith documented loads of 900 to 2000 pounds to produce femoral neck fractures in cadavers. Assuming that these specimens were probably from older individuals, one can extrapolate that even higher forces are needed to fracture the femoral neck in young adults. It is encouraging to note that appropriate treatment with urgent, accurate reduction and internal fixation can result in preservation of the patient’s hip in 85% of patients at 10 years of follow-up.
Certain conditions of abnormal bone metabolism significantly affect the diagnosis and treatment of femoral neck fractures. The following two conditions require individual consideration.
Because of the severe hip synovitis that can occur with rheumatoid arthritis, bone density is generally poor and chronic hip symptoms can mask acute femoral neck fractures. Williams and associates reported that four of five patients with rheumatoid arthritis and a femoral neck fracture had not fallen. This suggests that the majority of femoral neck fractures in patients with rheumatoid arthritis begin as fatigue fractures through severely osteoporotic bone. Femoral neck fractures can also occasionally be seen in an osteoarthritic hip, although patients with coxarthrosis generally have a lower risk of osteoporosis and femoral neck fracture. Treatment in both situations, because of the underlying articular disease, should generally be THA. This reliably restores such patients to their prefracture ambulatory status.
Patients undergoing chronic renal dialysis have metabolic bone disease and are prone to femoral neck fatigue fractures as outlined previously. This risk persists after renal transplantation. Patients with end-stage renal disease may not have higher mortality rates than age- and sex-matched controls; however, a hip fracture in a patient with end-stage renal disease does increase mortality risk. Displaced femoral neck fractures in chronic renal failure patients are generally treated best with THA. Internal fixation of displaced fractures in the setting of severe osteoporosis will be mechanically suboptimal. Biologic factors may interfere with healing as well.
Basilar Neck Fracture.
Fractures that occur in the lower part of the femoral neck are often called basilar neck fractures. However, there is no precise definition for this term. These injuries occur through an area of transition. Although generally true, it cannot be universally stated that these fractures are extracapsular and carry a better prognosis. A severely displaced low femoral neck fracture can still disrupt the lateral epiphyseal artery complex. From laboratory studies, Claffey determined this degree of displacement to be half the diameter of the femoral head superiorly (distal fragment relative to the femoral head). Because diagnostic radiographs do not show the extreme of displacement that occurs at the time of the injury, arterial disruption is a distinct possibility in the case of a basilar neck fracture. Although most surgeons favor an internal fixation construct that would be selected for an intertrochanteric hip fracture, the fracture should be viewed biologically as a femoral neck fracture (with its attendant risk of AVN of the femoral head) and treated as such with urgent fixation and capsulotomy. As with more proximal fractures, stable fixation is advisable to promote fracture healing.
Evolution of Treatment
Nonunion and AVN of the femoral head have long been recognized as the major problems associated with a femoral neck fracture. In 1901, Senn stated, “the only cause for nonunion in the case of an intracapsular fracture is to be found in our inability to maintain coaptation and immobilization of the fragments during the time required for bony union to take place.” He had previously advocated immediate reduction and internal fixation of these fractures in 1889 and had published animal data to support the concept that these fractures would heal with internal fixation. Whitman in 1902 and 1933 and Cotton in 1927 and 1934 advocated closed reduction and impaction followed by placement into a spica cast in internal rotation as the method of choice for the management of femoral neck fractures. Leadbetter further detailed the method of closed reduction and stated that “plaster fixation cannot be expected to yield good results consistently and logically in more than 65 or 75 percent of these [femoral neck fracture] cases.” Phemister outlined the pathophysiology of “creeping substitution” as it pertains to AVN of the femoral head after femoral neck fracture.
The first widely accepted method of internal fixation was reported by Smith-Petersen and colleagues in 1931. Use of their triflange nail was reported in many publications until the mid-1970s. The most highly documented series was that of Fielding and associates published in 1962, which revealed a nonunion rate of 18% and an AVN rate of 29% in a series of 284 displaced and nondisplaced femoral neck fractures. Moore published a report on the first implant with multiple pins (adjustable nails) in 1937. A 96% rate of union was reported. Several designs of implants with multiple pins followed shortly thereafter. Moore then developed a prosthetic replacement for the femoral head and reported its use in 33 cases in 1952. The Thompson prosthesis was developed shortly thereafter, and the current debate of whether to fix the femoral neck fracture or replace the femoral head began. Sliding devices such as the Pugh nail (1955) and the Richards screw (1964) were developed for controlled impaction of femoral neck fractures. More designs with multiple pins and screws were added to the already numerous devices in the late 1970s and 1980s. In Europe, 130-degree blade plates enjoyed fairly widespread acceptance. Now, however, when internal fixation is warranted, most surgeons use some type of multiple pins or screws, or a compression hip screw with an added screw or pin to prevent rotation of the femoral head. That quality of reduction is a key factor in achieving union has been reconfirmed repeatedly; placement of implants with multiple screws to within 3 mm of the inferior femoral neck cortex has also been shown to be important in achieving fracture union. Stereophotogrammetric analysis has confirmed the critical importance of fracture stability for promotion of fracture healing.
Judet developed a method of placing a viable bone graft across the posterior aspect of a femoral neck fracture and into the femoral head to decrease the incidence of AVN and nonunion. This quadratus femoris muscle pedicle graft was advocated by Meyers and colleagues, who reported an 8% incidence of posttraumatic AVN and an 11% incidence of nonunion. Subsequent reports have not confirmed these results, however.
Current Algorithm for Treatment of Femoral Neck Fracture
Preventing unnecessary motion of the injured hip may protect it from additional damage, as well as minimize the patient’s discomfort. Moderate flexion and external rotation increases the volume of the hip capsule and may thus decrease intracapsular pressure, possibly improving femoral head perfusion. A pillow under the knee will accomplish this. The leg should be elevated enough from the mattress to protect the heel from pressure and subsequent skin breakdown. Preoperative skin traction is ineffective for decreasing pain.
Undisplaced Femoral Neck Fractures.
As defined previously, undisplaced femoral neck fractures include both valgus impacted (Garden grade I) and complete (Garden grade II) femoral neck fractures because of the similar prognosis of both fracture types. Their treatment is the same. Internal fixation is indicated for most undisplaced acute femoral neck fractures . It has clearly been shown that mobilization of the patient results in a lower mortality rate. Internal fixation allows mobilization of the patient without loss of fracture reduction in most initially undisplaced cases. With conservative treatment (recumbent position for 7 weeks), the displacement rate or disimpaction rate has been shown to be 10% to 27% by Bentley, Hilleboe and colleagues, and Jensen and Hogh. Bentley reported that rates of AVN after a nondisplaced femoral neck fracture were 14% for conservative treatment and 18% for internal fixation. If displacement occurs, however, the rates more than double, and prosthetic replacement may become advisable for older patients. Therefore, internal fixation seems justified, with multiple pins or screws using an implant of the surgeon’s choice. No difference in clinical results is apparent with cannulated screws or Knowles pins. Poor placement of the implants is associated with nonunion and failure of fixation. Because insertion of a nail-type device might displace the fracture, nails should not be used. Several laboratory and clinical publications support the concept of performing a capsulotomy to release excessive pressure caused by fracture bleeding into the hip joint capsule. Recently, Rogmark and associates reported the largest series of 224 patients treated with internal fixation for nondisplaced femoral neck fractures. The failure rate was 11% with most having AVN. The secondary arthroplasty rate was 9%. One retrospective review suggests that the reoperation rate for nondisplaced or impacted femoral neck fractures in patients older than 80 years who are treated by internal fixation may warrant consideration of hemiarthroplasty. Parker and associates published a comparison of 346 patients with nondisplaced femoral neck fractures that had internal fixation and another cohort of 346 patients that had hemiarthroplasty for undisplaced femoral neck fractures. The conclusion was that internal fixation was better than hemiarthroplasty with regard to decreasing operating time, hospital stay, perioperative complications, and 1-year mortality. The internal fixation group had less pain at 1 year, less loss of mobility, and less dependence on assistive device for walking.
Displaced Femoral Neck Fracture.
As previously discussed, treatment of displaced femoral neck fractures is aimed at restoring hip function. Rapid mobilization of the patient is thought to reduce the risk of medical complications and improve the ultimate functional outcome. Additionally, it decreases the costly length of stay in an acute care hospital. Failure of fracture fixation, nonunion, and AVN with symptomatic late segmental collapse have long been recognized as serious complications that compromise the results of treatment of femoral neck fractures. In striving to provide mobilization while avoiding these and other complications, treatment for displaced femoral neck fractures has evolved from closed reduction and casting, to internal fixation, to prosthetic replacement, and presently to selective use of prosthetic replacement or internal fixation. The currently offered algorithm recommends internal fixation, after closed or open reduction, for most patients with adequate bone density. Level of the femoral neck fracture does not influence the rate of AVN or nonunion and generally should not be used to determine treatment. Prosthetic replacement is reserved for physiologically older patients in whom internal fixation is unlikely to succeed—those with marked osteopenia, fracture comminution, or both. In general, such patients are physiologically elderly, with low functional demands. Their ambulation is at best restricted to their domicile, they may be unable to assist with their own care, and their life expectancy is often limited. They are therefore at less risk of having late complications that might require revision of an arthroplasty. Although different types of prosthetic replacement for the proximal part of the femur have relative advantages and disadvantages, none can provide as durable and functional a hip as that regained by satisfactory bone healing. The DXA diagnoses of osteopenia or osteoporosis do not correlate with failure of internal fixation. Furthermore, failure after internal fixation of a femoral neck fracture can be satisfactorily salvaged by THA, which has a low rate of complications. Failed hemiarthroplasties require a similar procedure, though a more difficult one with possibly poorer results.
Treatment of Fatigue or Stress Fractures.
Considerations for treatment of fatigue fractures of the femoral neck are given in Table 54-6 .
|Type of Fracture||Treatment|
|Transverse type||Urgent internal fixation with multiple pins|
|Compression type||Mobilization with limited weight-bearing on crutches or a walker|
Pathologic Femoral Neck Fractures.
The reader is referred to Chapter 20 for details of the diagnosis and treatment of pathologic fracture of the femoral neck. If surgery is indicated, some form of arthroplasty is necessary.
Patients with Multiple Injuries.
Nearly all femoral neck fractures in young patients with normal bone are secondary to high-energy trauma. Fifty percent to 70% of patients younger than 50 years of age with a nonfatigue, displaced femoral neck fracture will have other organ system injuries. A young patient with a displaced femoral neck fracture should be assumed to have other injuries until proven otherwise.
Indications for Operative Intervention.
Surgical treatment is warranted for all but the most medically fragile and bedbound patients with displaced femoral neck fractures. Operative management has been shown to achieve significant functional benefit in cost-benefit analyses. The basic choice of treatment is between internal fixation and arthroplasty. This controversy is discussed in further detail later. It is recommended that several factors be considered when choosing treatment for patients with displaced femoral neck fractures.
Age alone is a poor predictor of activity level, bone quality, physiologic status, and life expectancy, all of which should also be considered when deciding between reduction and fixation and replacement arthroplasty. If successful, reduction (either closed or open) and internal fixation provide the best and most durable results. Failure is caused by early loss of fixation, by nonunion, or by symptomatic segmental collapse from AVN. Although not always predictable, fixation problems are most common in patients with osteopenic bone and comminution. Prosthetic replacement of the proximal part of the femur avoids the problems of nonunion and AVN but may be associated with higher perioperative morbidity than seen with internal fixation. In addition, arthroplasty poses late problems of loosening and acetabular erosion, either of which may require revision surgery. Results after revision of failed hemiarthroplasties are not as good as those after primary THA, and the procedure may be quite difficult. Therefore, initial treatment of a displaced femoral neck fracture should seek to minimize the likelihood of future revision for failed arthroplasty. It appears that this is best accomplished by restricting hemiarthroplasties to low-demand users with limited life spans. Improvement in implant design may increase their durability and ease of revision, although the data to support this assertion remain inconclusive.
Factors that suggest the advisability of prosthetic replacement include pathologic bone, severe chronic illness (especially rheumatoid arthritis, osteoarthritis, or chronic renal failure), and a significantly limited life span. Advanced chronologic age alone is a questionable indication for hemiarthroplasty. The average life expectancy for 75-year-olds is more than 10 years. Therefore, many orthopaedists would extend the indications for internal fixation into the early and mid-70s for active individuals with good bone density and no chronic illness. Inactive, osteoporotic elderly patients with a limited life span are candidates for simple unipolar hemiarthroplasties. Those with displaced femoral neck fractures who can ambulate functionally outside their homes (community ambulators) and whose likelihood of success with internal fixation is low should be considered candidates for hip arthroplasty, either total hip replacement or hemiarthroplasty, the latter generally being reserved for less active patients.
Timing of Surgery.
There is no consensus for advocating or delaying surgery for femoral neck fractures. Proponents of early surgery would suggest that prompt reduction of displaced femoral neck fractures will unkink the vessels and allow intracapsular decompression to remove the offending agent of increased intracapsular pressure Restoring the anatomy will improve blood flow to the femoral head, thus minimizing the risk of femoral head osteonecrosis. Swiontkowski and associates recommended that treatment of femoral neck fractures should be performed within 8 hours after injury. Subsequent studies also suggested that early surgery (within 6 to 12 hours) can decrease the rate of femoral head osteonecrosis.
Jain and associates published a series of 38 patients with subcapital hip fractures in a comparative cohort design. They compared early (<12 hours) and delayed (>12 hours) surgical fixation. Their results showed radiographic evidence of femoral head AVN in 16% of the patients (all in the delayed fixation group). There were no differences in the rate of AVN with regard to age, fracture displacement, and method of fracture fixation. Functional outcomes were collected and they did not find a difference in the functional results between the patients that developed osteonecrosis and the patients that did not have osteonecrosis. Delayed treatment had an increased rate of osteonecrosis, but did not affect the functional outcomes.
Subsequent studies published did not show any differences in the rate of osteonecrosis with delayed surgery greater than 24 hours. Upadhyay and associates reported a prospective, randomized study comparing ORIF and closed reduction and percutaneous pinning (CRIF) in young adults with displaced femoral neck fractures. Forty-four femoral neck fractures were randomized to the ORIF (a Watson-Jones approach with a T-shape incision in the capsules) and 48 femoral neck fractures were randomized to the CRIF. The majority of the patients were treated more than 48 hours after injury, however, which likely dampened the impact of the open reduction. The result showed no difference in the rate of osteonecrosis between the two groups (14.6% for the CRIF and 18.2% for the ORIF). They evaluated other risk factors (age, gender, time to surgery [<48 or >48 hours], and posterior comminution) and there was no correlation with the development of osteonecrosis.
Delay from fracture to hospital admission significantly increases the risk of mortality. Many older patients with femoral neck fractures have significant medical problems, which increase the risk of mortality. Correcting these problems preoperatively may greatly reduce the risk of complications. Therefore, extreme care should be taken to optimize the patient’s condition for surgery (see Chapter 53 ). Delay to surgery for “systems failures” (i.e., lack of operating room availability) should, however, be avoided. One risk-adjusted analysis did not reveal an increased association between mortality and delay to surgery. However, Zuckerman and colleagues showed increased mortality in patients with two or fewer comorbidities in whom surgery had been delayed for more than 2 calendar days. Parker and Pryor confirmed the increased morbidity in patients who have no comorbidities but in whom surgery is delayed. Similarly, Rogers and associates documented a significant increase in mortality in physiologically stable elderly patients with isolated proximal femoral fractures and a delay of more than 24 hours before surgery. Early surgery was not associated with improved function or mortality in one recent prospective chart study. Delay longer than it takes to stabilize patients medically is not advised.
In summary, young patients with femoral neck fractures should have the surgery performed on an urgent basis. Femoral neck fractures in young adults are often caused by high-energy injuries and may have other associated injuries. As soon as the patient is considered medically stable and cleared to undergo anesthesia, ORIF of the femoral neck should be done. As for the elderly patients with multiple medical conditions, it is best to optimize them medically before proceeding with surgical intervention. However, surgery should not be delayed more than 2 days. This may require careful coordination and communication among the patient’s health care providers (orthopaedic surgeon, hospitalist, anesthesiologist, nurse, and patient and/or family members).
Historically, more than 100 different internal fixation devices have been used for stabilizing femoral neck fractures. Triflanged nails, either with or without femoral shaft side plates, should not be used to internally fix femoral neck fractures. The distraction produced when using these devices has been shown to have an adverse effect on femoral head blood flow. The device chosen should have a mechanism for producing compression across the fracture site.
The recommended implant(s) for internal fixation of a femoral neck fracture are multiple pins or some type of cannulated or noncannulated cancellous screw. Screw cannulation allows the use of guide wires, thereby enabling the use of guide systems to help achieve parallel placement of the implants. In general, the amount of compression generated across the fracture is proportional to the thread area of the screw. The critical element in fracture stability is the density of the bone. The use of more than three implants does not seem to provide any increased mechanical advantage. Washers do add to the stability of fixation.
Brodetti has shown that large implants, if placed suboptimally in the posterior or superior aspect of the femoral head, can damage the blood supply to the femoral head. However, Ort and Lamont have reported results with compression screws that are equal to those of other techniques, and Bonnaire and associates reported that their results (in terms of osteonecrosis) were better with sliding hip screws. If this device is selected, a second pin or screw should be inserted superior to the guide wire for the large compression screw, to control rotation of the femoral head fragment. Because the hip compression screw by itself controls rotation poorly when compared with implants with multiple screws or pins, this superior screw should remain as part of the definitive fixation. Hernefalk and Messner demonstrated increased stiffness of implants with side plates. Malposition of the screw within the femoral head in patients with a femoral neck fracture has been shown to be more common in patients with left hip fractures when the surgeon is right hand dominant.
Recently, there are multiple biomechanical and clinical studies evaluating the different fixation methods for femoral neck fractures. For the majority of femoral neck fractures, fixation with multiple cancellous lag screws is recommended. Ideally, the screws are placed parallel to each other and perpendicular to the fracture line to obtain optimal compression at the fracture. Femoral neck fracture that are more transverse or horizontal (Pauwels type I and II fractures) are most amenable to this type of fixation. These three cancellous lag screws should be in an inverted triangle configuration. This apex-distal screw orientation has less risk of producing a subtrochanteric fracture when compared to the apex-proximal orientation. The distal screw should be placed along the medial femoral neck to resist varus displacement. The use of a fourth screw may be beneficial in femoral neck fractures where there is posterior comminution. However, a fourth screw does not increase in mechanical strength enough in most femoral neck fractures to justify its use. Two cannulated screws are inadequate fixation for a displaced femoral neck fracture.
The sliding hip screw (SHS) is another good implant of choice, especially for basicervical femoral neck fractures with comminution. SHS has been shown to provide more stable fixation than three cancellous screws for basicervical femoral neck fracture patterns. Blair and associates published a biomechanical cadaver study evaluating three different fixation techniques for treatment of basicervical femoral neck fracture. They recommend the SHS implant for this fracture pattern. Moreover, they suggested that a derotation screw located superior to SHS did not add any increase in fixation after the SHS is placed. However, a derotational screw can be beneficial to prevent rotation of the femoral head during insertion of the compression screw.
The femoral neck fracture with a more vertically oriented fracture pattern (Pauwels type III) can be very challenging to obtain and maintain reduction. This high-angle fracture pattern is associated with higher rates of failure and nonunion. The key to successfully treating this high-angle fracture pattern with three cancellous screws is obtaining an anatomic reduction and placing the screws strategically. This is best accomplished through an open approach to visualize the fracture, anatomic reduction of the fracture, and fracture compression with three screws, optimally placed in parallel. The first screw should be placed inferior along medial femoral neck on the AP imaging and centered on femoral neck in the lateral imaging. The second should be placed superiorly on the AP imaging and anterior along the neck in the lateral imaging. The third screw should be placed superiorly on the AP imaging and posterior along the neck in the lateral imaging. A good spread, if not divergent, placement of the three screws on the lateral view is important ( Fig. 54-17 ). This will help maintain the reduction and decrease the risk for nonunion.
A SHS is another good option for Pauwels type III femoral neck fractures ( Fig. 54-18 ). Baitner and associates compared SHS with three cannulated cancellous screws for the Pauwels type III fractures. They concluded that fixation with SHS resulted in less inferior femoral head displacement, less shearing displacement, and a greater load to failure. Bonnaire and Weber evaluated four different methods of fixation for Pauwels type III fractures using a cadaveric model. They compared SHS with a derotational screw, SHS without derotational screw, cancellous screws, and a 130-degree blade plate. Their results showed that SHS with the derotational screw is the best implant for this fracture pattern. However, there are disadvantages to using these large compression hip screws. There are concerns for the amount of bone loss if subsequent reconstruction is required for nonunion, the risk of disrupting blood supply to the femoral head if imperfectly placed (particularly posterior-superior position), and their inability to adequately control rotation without inserting an additional derotational screw.
Fixation of femoral neck fracture with multiple cancellous screws allows for excellent compression but does not provide length stability. Zlowodzki and associates reported a shortening rate of 31% for undisplaced and 27% for displaced femoral neck fractures. Femoral neck shortening does have a negative effect on the functional capacity and quality of life. Aminian and associates evaluated the biomechanical stability of the fixed-angle proximal femoral locking plate (PFLP), the 7.3-mm cannulated screws, the 135-degree dynamic hip screw (DHS), and the 95-degree dynamic condylar screw (DCS) for fixation of femoral neck fractures. Using cadaveric femurs and creating a Pauwels type III fracture, they found that the strongest construct was the PFLP. This is followed by DCS, DHS, and the three-cannulated-screw construct. One of the advantages of the PFLP is that it allows for multiple fixed-angle points of fixation into the femoral head. The major disadvantage is that proper anatomic reduction and compression of the fracture is necessary prior to fixation, as the PFLP does not allow for fracture compression. Boraiah and associates published their series of 54 femoral neck fractures that were treated with length-stable implants after intraoperative fracture site compression. Fifty-one (94%) patients healed their femoral neck fracture with minimal shortening. In contrast, Berkes and associates reported their clinical series of 18 patients who were treated with posterolateral femoral locking plate (PLFLP). The purpose of using the locking plate construct was to provide length-stable rigid fixation and prevent femoral neck collapse and shortening. The result of this study showed a poor outcome with failure in seven patients. The authors suggested that the construct was too stiff to allow for small fracture site micromotion that would encourage bony union. Most of the failures occurred with screw breakage and resulted in varus collapse. There is insufficient clinical experience with the PFLP to recommend its routine use.
Role of Capsulotomy.
Intracapsular tamponade has similarly been found to have a negative influence on femoral head blood flow in numerous laboratory and clinical studies. Performing an anterior hip capsulotomy at the time of surgical fixation remains controversial. There are both animal and clinical studies that support the role of capsulotomy. Animal studies have shown reduced blood flow to the femoral head when there is an increase in intracapsular hip pressure. Multiple clinical studies have evaluated the impact of increased intracapsular pressure in nondisplaced femoral neck fractures. Even nondisplaced femoral neck fractures can have intracapsular hip pressure that exceed the normal hip pressure. Reducing the hip pressure can be done with hip aspiration or better with open or percutaneous capsulotomy. The theoretical advantage of decreasing the intracapsular pressure will lead to improved blood flow to the femoral head. Removal of this tamponade effect may reduce femoral head ischemia and ultimately decrease the risk of femoral head AVN. There are recent clinical series that showed AVN occurred even in the nondisplaced femoral neck fractures. It is difficult to conceive of a mechanism for posttraumatic femoral head osteonecrosis apart from intracapsular tamponade in a nondisplaced fracture. Moreover, a recent study by Papakakis and associates showed that the posterior retinaculum of the hip remains intact in most displaced femoral neck fractures. They prospectively evaluated 112 patients with Garden type III and IV femoral neck fractures that underwent hemiarthroplasty. Intraoperatively, the anterior and posterior retinacula of the hip were inspected for tears. They found that the posterior retinaculum was intact in all Garden type III and in 95.2% of completely displaced femoral neck fractures. This study advocated the benefit of performing an anterior capsulotomy for displaced femoral neck fracture treated with internal fixation.
Although some authors disagree, the current recommendation is to extend the standard lateral approach into the Watson-Jones interval between the tensor fasciae latae and gluteus medius and perform an anterior capsulotomy under direct vision ( Fig. 54-19 ). Although possibly only 5% to 15% of fractures are seen in which the hip capsule is not disrupted and sufficient intracapsular pressure has accumulated to impede venous drainage or arteriolar supply of the femoral head, these fractures can be effectively treated by this simple maneuver. Capsulotomy adds 5 to 10 minutes to the dissection and exposes the patient to no further risk. There are no published reports of complications with performing an open anterior capsulotomy (directly visualizing the capsule and performing a capsulotomy). However, detachment of the scalpel blade from the knife handle during a percutaneous capsulotomy has been reported ( Fig. 54-20 ). After successful closed reduction and stable fixation of the femoral neck, we recommend performing a percutaneous capsulotomy with a No. 10 or 15 blade ( Fig. 54-21 ). To avoid detachment of the blade, the surgeon should make sure the blade is fully seated on the knife handle and can wrap the blade and the handle with a narrow strip of Ioban (3M, St. Paul, MN). The blade is slid over the anterior trochanter and in line with the center of femoral neck on AP C-arm images. Lateral C-arm images should be obtained and the capsulotomy should be performed in the lateral view (making sure you are right on top of the femoral neck bone by feel and by C-arm images). If there is a mini-incision made to place the implants, then a flash of hematoma should be seen to ooze out when the capsulotomy is complete. Furthermore, anatomic reduction can then be achieved under direct vision. In delayed treatment of displaced femoral neck fractures, open reduction does not favorably impact the rate of AVN.