Nonunions: Evaluation and Treatment


Introduction

Fracture nonunions may represent a very small percentage of the traumatologist's case load but can account for a high percentage of a surgeon's stress, anxiety, and frustration. A fracture nonunion may be anticipated after a severe traumatic injury, such as an open fracture with segmental bone loss, but may also appear after a low-energy fracture that seemed destined to heal.

Fracture nonunion is a chronic medical condition associated with pain and functional and psychosocial disability. Two large studies including tibial ( n = 243) and femoral ( n = 188) nonunions reported severe effects of nonunion on SF-12 Physical Component Summary scores, Mental Component Summary scores, Brief Pain Inventory, and American Academy of Orthopaedic Surgeons (AAOS) Lower Limb Core Scale. The impact on physical health was comparable to end-stage hip arthrosis and exceeded that of many chronic health conditions, including congestive heart failure, and pain and limb-specific disability were both severe and interfered with function and quality of life. Another study of 832 long bone nonunions reported an average time trade-off–based utility score (where 0 = death and 1.00 = perfect health) of 0.68, substantially worse than that reported for type 1 diabetes mellitus, stroke, and acquired immunodeficiency syndrome.

Some 90% to 95% of all fractures heal without problems. Nonunions are the small percentage of cases in which the biologic process of fracture repair cannot overcome the local biology and mechanics of the bony injury.

Definitions

A fracture is said to have “gone on to nonunion” when the normal biologic healing processes cease to the extent that solid healing will not occur without further treatment intervention. The definition is subjective, with criteria that result in high interobserver variability.

The literature reveals myriad definitions of nonunion. For the purposes of clinical investigations, the US Food and Drug Administration (FDA) defines a nonunion as a fracture that is at least 9 months old and has not shown any signs of healing for 3 consecutive months. Müller's definition is the failure of a (tibia) fracture to unite after 8 months of nonoperative treatment. These two definitions are widely used, but their arbitrary use of a temporal limit is flawed. For example, several months of observation are not required to declare a tibial shaft fracture with 10 cm of segmental bone loss to be a nonunion (i.e., an injury that will not heal without further intervention). Conversely, how does one define a fracture that continues to consolidate but requires 12 months to heal completely?

We define nonunion as a fracture that, in the opinion of the treating physician, has a zero possibility of healing without further intervention. We define delayed union as a fracture that, in the opinion of the treating physician, shows slower progression to healing than anticipated and is at risk of nonunion without further intervention.

To understand the biologic processes and clinical implications of fracture nonunion, an understanding of the normal fracture healing process is required.

Fracture Repair

Fracture repair is an astonishing process that involves spontaneous, structured regeneration of bony tissue and restoration of mechanical stability. The process begins at the moment of bony injury, initiating a proliferation of tissues that ultimately leads to healing.

The early biologic response at the fracture site is an inflammatory response with bleeding and the formation of a fracture hematoma. The repair response is initiated by the presence of osteoprogenitor cells from the periosteum and endosteum and hematopoietic cells capable of secreting growth factors. After solid fracture healing, bone remodeling progresses according to Wolff's law.

The repair process, involving both intramembranous and enchondral bone formation, requires mechanical stability, an adequate blood supply, good bony contact, and the appropriate endocrine and metabolic responses. The type of healing is related to the extent of injury and the type of treatment ( Table 26.1 ).

Table 26.1
Type of Fracture Healing Based on Type of Stabilization
Type of Stabilization Predominant Type of Healing
Cast (closed treatment) Periosteal bridging callus and interfragmentary enchondral ossification
Compression plate Primary cortical healing (cutting cone-type remodeling)
Intramedullary nail Early: periosteal bridging callus
Late: medullary callus
External fixator Dependent on extent of rigidity
Less rigid: periosteal bridging callus
More rigid: primary cortical healing
Inadequate immobilization with adequate blood supply Hypertrophic nonunion (failed enchondral ossification)
Inadequate immobilization without adequate blood supply Atrophic nonunion
Inadequate reduction with displacement at the fracture site Oligotrophic nonunion

Healing via Callus

In the absence of rigid fixation (e.g., during cast immobilization), stabilization of bone fragments occurs by periosteal and endosteal callus formation. If the fracture site has an adequate blood supply, callus forms and increases the cross-sectional area at the fracture surface, which enhances fracture stability. Formation of fibrocartilage provides further fracture stability, replacing granulation tissue at the fracture site. Enchondral bone formation, in which bone replaces cartilage, occurs after calcification of the fibrocartilage.

Direct Bone (Osteonal) Healing

Direct osteonal healing occurs without external callus and is characterized by the gradual disappearance of the fracture line over time. This process requires an adequate blood supply and absolute rigidity at the fracture site, most commonly accomplished via compression plating. In areas of direct bone-to-bone contact, fracture repair resembles cutting-cone type remodeling. In areas where small gaps exist between fracture fragments, “gap healing” occurs via appositional bone formation.

Indirect Bone Healing

Indirect bone healing occurs in fractures that have been stabilized using methods providing less-than-absolute rigidity, including intramedullary nail fixation, tension band wire techniques, cerclage wiring, external fixation, and plate-and-screw fixation (when applied suboptimally). Indirect healing involves coupled bone resorption and bone formation at the fracture site. Healing occurs via a combination of external callus formation and enchondral ossification.

Etiology of Nonunions

Predisposing Factors—Instability, Inadequate Vascularity, Poor Contact

The most basic requirements for fracture healing include (1) mechanical stability, (2) adequate blood supply (i.e., bone vascularity), and (3) bone-to-bone contact. The absence of one or more of these factors predisposes the fracture to the development of a nonunion. The factors may be negatively affected by the severity of the injury, suboptimal surgical fixation, or a combination of injury severity and suboptimal surgical fixation.

Instability

Mechanical instability can follow internal or external fixation. Factors producing mechanical instability include inadequate fixation (implants too small or too few), distraction of the fracture surfaces (hardware is as capable of holding bone apart as holding bone together), bone loss (segmental defect), and poor bone quality (i.e., poor purchase; Fig. 26.1 ). An adequate blood supply with excessive motion at the fracture site results in abundant callus formation, widening of the fracture line, failure of fibrocartilage to mineralize, and ultimately failure to unite.

Fig. 26.1, Mechanical instability at the fracture site can lead to nonunion. Mechanical instability can be caused by the following: (A) Inadequate fixation: A 33-year-old man presented with a femoral shaft nonunion 8 months after inadequate fixation with flexible intramedullary nails. (B) Distraction: A 19-year-old man with a tibia fracture treated with plate-and-screw fixation; this patient is at risk for nonunion because of distraction at the fracture site. (C) Bone loss: A 57-year-old man presented with segmental bone loss after débridement of a high-energy open tibia fracture. (D) Poor bone quality: A 31-year-old woman presented 2 years after open reduction internal fixation for an ulna shaft fracture; loss of fixation proved to be due to poor bone from chronic osteomyelitis.

Inadequate Vascularity

Loss of blood supply to the fracture surfaces may result from injury severity or surgical dissection. The extent of soft tissue injury is related to the risk of fracture nonunion. Open fractures and high-energy closed injuries may strip soft tissues, damage the periosteal blood supply, and disrupt the nutrient vessels, impairing the endosteal blood supply.

Injury of certain vessels, such as the posterior tibial artery, may also increase the risk of nonunion. Vascularity may also be compromised by excess stripping of the periosteum as well as damage to bone and the soft tissues during open reduction and hardware insertion. Whatever the cause, inadequate vascularity results in necrotic bone at the ends of the fracture fragments, which often results in fracture nonunion.

Poor Bone Contact

Poor bone-to-bone contact at the fracture site may result from soft tissue interposition; bone loss; and malposition, malalignment, or distraction of fracture fragments (see Fig. 26.1 ). Poor bone contact compromises mechanical stability and creates a defect.

The probability of fracture union decreases as defects increase in size. The threshold value for rapid bridging of cortical defects via direct osteonal healing, the so-called osteoblastic jumping distance, is approximately 1 mm in rabbits but varies from species to species. Larger cortical defects may also heal, but at a much slower rate, and bridge via woven bone. The “critical defect” represents the distance between fracture surfaces that will not be bridged by bone without intervention. The critical defect size depends on a variety of injury-related factors and varies considerably among species.

Other Contributing Factors

In addition to mechanical instability, inadequate vascularity, and poor bone contact, other factors may contribute to the development of nonunion ( Box 26.1 ). These factors are not direct causes of nonunion, per se.

Box 26.1
Etiology of Nonunions

Predisposing Factors

Mechanical Instability

  • 1.

    Inadequate fixation

  • 2.

    Distraction

  • 3.

    Bone loss

  • 4.

    Poor bone quality

Inadequate Vascularity

  • 1.

    Severe injury

  • 2.

    Excessive soft tissue stripping

  • 3.

    Vascular injury

Poor Contact

  • 1.

    Soft tissue interposition

  • 2.

    Malposition or malalignment

  • 3.

    Bone loss

  • 4.

    Distraction

Contributing Factors

  • Infection

  • Nicotine/cigarette smoking

  • Certain medications

  • Advanced age

  • Systemic medical conditions

  • Poor functional level

  • Venous stasis

  • Burns

  • Radiation

  • Obesity

  • Alcohol abuse

  • Metabolic and endocrine abnormalities

  • Malnutrition

  • Vitamin deficiencies

Infection

Infection in the zone of fracture increases the risk of nonunion. Infection may result in fracture-site instability as implants loosen in infected bone. Avascular, necrotic bone at the fracture site (sequestrum), common with infection, discourages bony union. Infection also produces poor bony contact as ingrowth of infected granulation tissue results in osteolysis.

Nicotine and Cigarette Smoking

Cigarette smoking adversely affects fracture healing. Nicotine inhibits vascular ingrowth and early revascularization of bone and diminishes osteoblast function. In rabbit models, cigarette smoking and nicotine impair bone healing in fractures, in spinal fusion, and during tibial lengthening.

Patients who smoke have higher rates of delayed fracture healing and nonunion. Cigarette smoking is also associated with osteoporosis and generalized bone loss, so mechanical instability due to poor bone quality for purchase may play a role.

Certain Medications

A number of animal studies have shown that nonsteroidal antiinflammatory drugs (NSAIDs) negatively affect the healing of experimentally induced fractures and osteotomies. Other animal studies have reported no significant effect.

Delayed long bone fracture healing has been documented in humans taking oral NSAIDs, although the mechanism of action (direct action at the fracture site vs. indirect hormonal actions) remains obscure. Whether all NSAIDs display similar effects and the dose–response characteristics of specific NSAIDs relative to fracture healing remain unknown.

Other medications have been postulated to affect fracture healing adversely, including phenytoin, ciprofloxacin, steroids, anticoagulants, and others.

Additional Contributing Factors

Other factors that may contribute to fracture nonunion include advanced age, systemic medical conditions (e.g., diabetes), poor functional level with the inability to bear weight, venous stasis, burns, radiation, obesity, alcohol abuse, metabolic bone disease, malnutrition and cachexia, and vitamin deficiencies.

Rodent studies have shown that albumin deficiency produces a fracture callus with reduced strength and stiffness, although early fracture healing proceeds normally. Dietary supplementation of protein during fracture repair reverses these effects and augments fracture healing. Protein intake in excess of normal daily requirements is unnecessary.

The presence of multiple contributing factors increases the risk of nonunion. Key factors include type IIIB injury, compartment syndrome, chronic medical conditions, open fracture, male sex, American Society of Anesthesiologists (ASA) Physical Status score, and percentage of cortical contact at the time of initial fracture fixation. Less than 25% cortical continuity, fracture gap after fixation, open fracture, and a transverse fracture pattern were reported to be predictive of risk of nonunion in a review of 200 tibial fractures.

Evaluation of Nonunions

No two patients with fracture nonunion are identical. The evaluation process is perhaps the most critical step in the patient's treatment pathway and is when the surgeon begins to form opinions about how to heal the nonunion. The goals of the evaluation are to discover the etiology of the nonunion and form a plan for healing the nonunion. Without understanding the etiology, the treatment strategy cannot be based on knowledge of fracture biology. A worksheet is an excellent method of assimilating the various data ( Fig. 26.2 ).

Fig. 26.2, Worksheet for patients with nonunions. NSAID, Nonsteroidal antiinflammatory drug; ROM, range of motion; VAS, visual analog scale.

Patient History

Evaluation begins with a thorough history, including the date and mechanism of the initial fracture, preinjury medical problems, disabilities, associated injuries, and pain and functional limitations related to the nonunion. The specific details of each prior surgical procedure to treat the fracture and fracture nonunion must be obtained through the patient and family, the prior treating surgeons, and a review of all medical records since the time of the initial fracture.

Knowledge of prior operative procedures is critical for designing the right treatment plan. Ignorance of any prior surgical procedure can lead to needlessly repeating surgical procedures that have failed to promote bony union in the past. Worse yet, ignorance of prior surgical procedures can lead to avoidable complications. For example, awareness of prior external fixation is important when the use of intramedullary nail fixation is contemplated because of an increased risk of infection.

The hospital records and operative reports from the time of the initial fracture may also be used to determine the condition of the tissues in the injury zone (open wounds, contamination, crush injuries, periosteal stripping, devitalized bone fragments, etc.) and the history of prior soft tissue coverage procedures.

The history should also include:

  • Details regarding prior wound infections, including culture reports in prior medical records;

  • Intravenous and oral antibiotic use, particularly if the patient remains on antibiotics at the time of presentation;

  • Problems with wound healing and episodes of soft tissue breakdown;

  • Other perioperative complications (venous thrombosis, nerve or vessel injuries, etc.); and

  • Prior use of adjuvant nonsurgical therapies, such as electromagnetic field and ultrasound therapy.

Finally, the patient should be questioned regarding other contributing factors for nonunion (see Box 26.1 ). The history of NSAID use should be obtained, and their use should be discontinued. The pack-year history of cigarette smoking should be documented, and active smokers should be offered a program to halt the addiction, although it is unrealistic to delay treatment of a symptomatic nonunion until the patient stops smoking.

Physical Examination

The general health and nutritional status of the patient should be assessed because malnutrition and cachexia diminish fracture repair. Arm muscle circumference is the best indicator of nutritional status. Obese patients with nonunions have unique management problems related to achieving mechanical stability in the presence of high loads and large soft tissue envelopes.

The skin and soft tissues in the fracture zone should be inspected. Active drainage, sinus formation, and deformity should be noted. A neurovascular examination can identify vascular insufficiency and motor or sensory dysfunction.

The nonunion site should be manually stressed to evaluate motion and pain. Assessment of nonunion site motion is difficult in limbs with paired bones in which one of the bones remains intact. Generally, nonunions with little or no clinically apparent motion have some callus formation and good vascularity at the fracture surfaces. Nonunions that display motion typically have poor callus formation, but may have vascular or avascular fracture surfaces.

Active and passive motion of the joints adjacent to the nonunion, both proximal and distal, should be performed. Motion at the nonunion site is frequently associated with diminished motion at an adjacent joint. For example, patients with a long-standing distal tibial nonunion often have a fixed equinus contracture and limited ankle motion ( Fig. 26.3 ). Similarly, patients with supracondylar humeral nonunions commonly have fibrous ankylosis of the elbow ( Fig. 26.4 ). Such problems may alter both the treatment plan and the expectations for the ultimate functional outcome.

Fig. 26.3, A 20-year-old woman presented with a distal tibial nonunion 22 months status post–high-energy open fracture. (A) Clinical photograph and (B) lateral radiograph showing apex anterior angulation at the fracture site resulting in the clinical equivalent of a severe equinus contracture.

Fig. 26.4, (A) Anterior-posterior (AP) radiograph of a 32-year-old man who presented with a supracondylar humeral nonunion. On physical examination, it can be difficult to differentiate motion at the nonunion site and the elbow joint. This patient had very limited range of motion at the elbow but gross motion at the nonunion site. Cineradiography can be useful for evaluating the contribution of the adjacent joint and the nonunion site to the arc of motion. Cineradiography showing flexion (B) and extension (C) of the elbow reveals that most of the motion is occurring through the nonunion site, not the elbow joint. This patient should be counseled preoperatively regarding elbow stiffness after stabilization of the nonunion.

Patients who have a stiff nonunion with an angular deformity may present with a compensatory fixed deformity at an adjacent joint. The fixed deformity at the joint must be recognized preoperatively, and the treatment plan must include its correction. Realignment of a stiff nonunion with a deformity without addressing an adjacent compensatory fixed joint deformity results in a straight bone with a deformed joint, thus producing a disabled limb.

For example, patients who have a stiff distal tibial nonunion with a varus deformity often develop a compensatory valgus deformity at the subtalar joint to achieve a plantigrade foot for gait. On visual inspection, the distal limb segment appears aligned, but radiographs show the distal tibial varus deformity. To determine whether the subtalar joint deformity is fixed or reducible, the patient is asked to position the subtalar joint in varus (i.e., invert the foot). If the patient cannot invert the subtalar joint, and the examiner cannot passively invert the subtalar joint, the joint deformity is fixed. Deformity correction will therefore be required at both the nonunion site and the subtalar joint. Conversely, if the patient can achieve subtalar inversion, the deformity at the joint will resolve with deformity correction at the nonunion.

In general, if the patient cannot place the joint into the position that parallels the deformity at the nonunion site, the joint deformity is fixed and requires correction. If the patient can achieve the position, the joint deformity will resolve with realignment of the long bone deformity ( Fig. 26.5 ).

Fig. 26.5, Angular deformity at a nonunion site, which is near a joint, can result in a compensatory deformity through a neighboring joint. For example, coronal plane deformities of the distal tibia can result in a compensatory coronal plane deformity of the subtalar joint. A deformity of the subtalar joint is (A) fixed if the patient's foot cannot be positioned into the deformity of the distal tibia or (B) flexible if it can be positioned into the deformity of the distal tibia. Sagittal plane deformities of the distal tibia can result in a sagittal plane deformity of the ankle joint. A deformity of the ankle joint is (C) fixed if the patient's foot cannot be positioned into the deformity of the distal tibia or (D) flexible if it can be positioned into the deformity of the distal tibia.

If bone grafting is contemplated, the anterior and posterior iliac crests should be examined for evidence (e.g., incisions) of prior surgical harvesting. For a patient who has had prior spinal surgery, determining which posterior crest has already been harvested may require plain radiographs or computed tomography (CT) scanning of the posterior iliac crests.

Radiologic Examination

Plain Radiographs

A review of the original fracture films reveals the character and severity of the initial bony injury. Comparison to the most recent plain radiographs can also show the progress or lack of progress toward healing.

Radiographs of the salient aspects of previous treatments will always tell the story of the nonunion to the astute observer. The complete series of prior plain films should be examined to determine:

  • The status of orthopaedic hardware (e.g., loose, broken, inadequate in size or number of implants) including removal or insertion on sequential films;

  • The evolution of deformity at the nonunion site—for example, whether a gradual process or single event;

  • The presence of healed or unhealed articular, butterfly, and wedge fragments; and

  • The time course of missing or removed bony fragments, added bone graft, and implanted bone stimulators for evaluation of the healing response to each.

The nonunion is next evaluated with a series of new radiographs:

  • An anterior-posterior (AP) and lateral radiograph of the involved bone, including the proximal and distal joints

  • AP, lateral, and two oblique views of the nonunion site on small cassette films to improve magnification and resolution ( Fig. 26.6 )

    Fig. 26.6, A nonunion is visualized better on small cassette views than on large cassette views (see Fig. 26.7 for comparison).

  • Bilateral AP and lateral 51-inch alignment radiographs for lower extremity nonunions for assessing length discrepancies and deformities ( Fig. 26.7 )

    Fig. 26.7, A 60-year-old man presented with a tibial nonunion and an oblique plane angular deformity as seen on the (A) 51-inch anterior-posterior (AP) alignment view and (B) 51-inch lateral alignment view.

  • Flexion/extension lateral radiographs to determine the arc of motion and to assess the relative contributions of the joint and the nonunion site to that arc

The current plain films are used to evaluate the following radiographic characteristics of a nonunion:

  • Anatomic location,

  • Healing effort and bone quality,

  • Surface characteristics,

  • Status of previously implanted hardware, and

  • Deformities.

Anatomic Location

Diaphyseal nonunions involve primarily cortical bone, whereas metaphyseal and epiphyseal nonunions largely involve cancellous bone. Intraarticular extension of the nonunion should also be assessed.

Healing Effort and Bone Quality

Assessment of radiographic healing effort and bone quality helps define the biologic and mechanical etiologies of the nonunion.

Assessment of Healing Effort: Assessment of Bone Quality:
Radiolucent lines Sclerosis
Radiolucent bone gaps Atrophy
Callus formation Osteopenia
Bony defects

Radiolucent lines along fracture surfaces suggest regions that are devoid of bony healing. The simple presence of radiolucent lines on plain radiographs is not synonymous with nonunion, just as the lack of a clear radiolucent line does not confirm fracture union ( Fig. 26.8 ).

Fig. 26.8, A definitive decision cannot always be made about bony union based on plain radiographs. (A) This 88-year-old woman is 14 months status post–distal tibial fracture treated elsewhere with external fixation. Anterior-posterior (AP) and lateral radiographs are shown. Is this fracture healed, or is there a nonunion? (See Fig. 26.17 .) (B) This 49-year-old man presented 13 months status post–open reduction internal fixation of a distal tibia fracture. AP and lateral radiographs are shown. Is this fracture healed, or is there a nonunion? (See Fig. 26.17 .)

Callus formation occurs in fractures and nonunions that have an adequate blood supply but does not necessarily imply the bone is solidly uniting. AP, lateral, and oblique radiographs should be assessed for callus bridging the injury zone. All radiographs should be carefully checked for radiolucent lines so that a nonunion with abundant callus is not mistaken for a solidly united fracture (see Fig. 26.8 ).

Weber and Cech have classified nonunions based on radiographic healing effort and bone quality into viable nonunions, which are capable of biologic activity, and nonviable nonunions, which are incapable of biologic activity.

Viable nonunions include hypertrophic nonunions and oligotrophic nonunions . Hypertrophic nonunions possess adequate vascularity and display callus formation. They are caused by inadequate mechanical stability allowing persistent motion at the fracture surfaces.

The fracture site is progressively resorbed with the accumulation of unmineralized fibrocartilage and displays a gradually widening radiolucent line with sclerotic edges. Capillaries and blood vessels invade both sides of the nonunion but do not penetrate the fibrocartilaginous tissue ( Fig. 26.9 ). As motion persists at the nonunion site, endosteal callus may accumulate and seal off the medullary canal, increasing the production of hypertrophic periosteal callus. Hypertrophic nonunions may be classified as elephant-foot type, with abundant callus formation, or horse-hoof type, with less abundant callus formation.

Fig. 26.9, Microangiogram of a hypertrophic delayed union of a canine radius. Note the tremendous increase in local vascularity. The capillaries, however, are unable to penetrate the interposed fibrocartilage (arrows).

Oligotrophic nonunions have an adequate blood supply but little or no callus formation and arise from inadequate reduction with displacement at the fracture site.

Nonviable nonunions have inadequate vascularity, which precludes the formation of periosteal and endosteal callus, and are incapable of biologic activity. The radiolucent gap observable on plain radiographs is bridged by fibrous tissue that has no osteogenic capacity.

An atrophic nonunion is the most advanced type of nonviable nonunion. Classically, the ends of the bony surfaces have been thought to be avascular, although some studies suggest otherwise. Radiographically, the fracture surfaces appear partially absorbed and osteopenic. Severe cases may have large sclerotic avascular bone segments or segmental bone loss.

Surface Characteristics

A nonunion's surface characteristics ( Fig. 26.10 ) are prognostic in regard to its resistance to healing with various treatment strategies. Surface characteristics evaluated on plain radiographs include:

  • the surface area of adjacent fragments,

  • the extent of current bony contact,

  • the orientation of the fracture lines (shape of the bone fragments), and

  • stability to axial compression (fracture surface orientation and comminution).

The nonunions that are simplest to treat have good bony contact and large, transversely oriented surfaces that are stable to axial compression.

Fig. 26.10, Surface characteristics at the site of nonunion. (A) Surface area. (B) Bone contact. (C) Fracture line orientation. (D) Stability to axial compression.

Status of Previously Implanted Hardware

Plain radiographs reveal the status and stability of previously implanted hardware and the mechanical construct used to fixate the bone. Loose or broken implants denote instability at the nonunion site (i.e., the race between bony union and hardware failure has been lost) that requires further stabilization before union can occur. Radiographs are also useful for planning hardware removal needed to carry out the next treatment plan.

Deformities

Having assessed for clinical deformity via physical examination, plain radiographs are used to characterize further and more fully all associated deformities. Deformities are characterized by location (diaphyseal, metaphyseal, epiphyseal), magnitude, and direction and are described in terms of length, angulation, rotation, and translation .

Deformities involving length include shortening and overdistraction, measured in centimeters on plain radiographs in comparison to the contralateral normal extremity and using a radiograph marker to correct for magnification. Shortening may result from bone loss (from the injury or débridement) or overriding fracture fragments (malreduction). Overdistraction may result from a traction injury or improper positioning at the time of surgical fixation.

Deformities involving angulation are characterized by the magnitude and direction of the apex of angulation . Pure sagittal or coronal plane deformities are simple to characterize. Coronal plane angulation in the lower extremity commonly results in mechanical axis deviation of the extremity ( Fig. 26.11 ). Varus deformities result in medial mechanical axis deviation, and valgus deformities result in lateral mechanical axis deviation.

Fig. 26.11, (A) In this example, there is a nonunion of the diaphysis of the tibia with a varus deformity resulting in medial mechanical axis deviation. Note the divergence of the anatomic axis of the proximal and distal fragments of the tibia. (B) Close-up view of a section of an anterior-posterior (AP) 51-inch alignment radiograph of a 37-year-old woman with an 18-year history of a tibial nonunion. Note the medial mechanical axis deviation (MAD) of 26 mm. (C) AP and lateral radiographs show a 25-degree varus deformity and a 21-degree apex anterior angulation deformity, respectively. (D) Characterization of the oblique plane angular deformity using the trigonometric method. (E) Characterization of the oblique plane angular deformity using the graphic method.

Oblique plane angular deformities occur in a plane that is neither the sagittal nor the coronal plane and can be characterized using either the trigonometric method or the graphic method (see Fig. 26.11 ).

Angulation at a diaphyseal nonunion is usually obvious on plain radiographs as divergence of the anatomic axes (mid-diaphyseal lines) of the proximal and distal fragments (see Fig. 26.11 ). The magnitude and direction of angulation can be measured by drawing the anatomic axes of the proximal and distal segments (see Fig. 26.11 ).

Angular deformities associated with nonunions of the metaphysis and epiphysis (juxtaarticular deformities) are less obvious and more challenging to evaluate because they cannot be characterized using the mid-diaphyseal line method. Recognition and characterization of a juxtaarticular deformity require using the angle formed by the intersection of a joint orientation line and the bone's anatomic or mechanical axis ( Fig. 26.12 ). When the angle formed differs markedly from the contralateral normal extremity, a juxtaarticular deformity is present. If the contralateral extremity is also abnormal (e.g., bilateral injuries), published normal values are used for evaluation ( Table 26.2 ).

Fig. 26.12, In this example, there is a nonunion of the proximal tibia with a valgus deformity, resulting in lateral mechanical axis deviation. The proximal medial tibial angle of 94 degrees is abnormally high compared with both the contralateral normal extremity and the population normal values (see Table 26.3 ).

Table 26.2
Normal Values Used to Assess Lower Extremity Metaphyseal and Epiphyseal Deformities (Juxtaarticular Deformities) Associated With Nonunions
Anatomic Site of Deformity Plane Angle Description Normal Values (in Degrees)
Proximal femur Coronal Neck shaft angle Defines the relationship between the orientation of the femoral neck and the anatomic axis of the femur. 130
(range, 124–136)
Anatomic medial proximal femoral angle Defines the relationship between the anatomic axis of the femur and a line drawn from the tip of the greater trochanter to the center of the femoral head. 84
(range, 80–89)
Mechanical lateral proximal femoral angle Defines the relationship between the mechanical axis of the femur and a line drawn from the tip of the greater trochanter to the center of the femoral head. 90
(range, 85–95)
Distal femur Coronal Anatomic lateral distal femoral angle Defines the relationship between the distal femoral joint orientation line and the anatomic axis of the femur. 81
(range, 79–83)
Mechanical lateral distal femoral angle Defines the relationship between the distal femoral joint orientation line and the mechanical axis of the femur. 88
(range, 85–90)
Sagittal Anatomic posterior distal femoral angle Defines the relationship between the sagittal distal femoral joint orientation line and the mid-diaphyseal line of the distal femur. 83
(range, 79–87)
Proximal tibia Coronal Mechanical medial proximal tibial angle Defines the relationship between the proximal tibial joint orientation line and the mechanical axis of the tibia. 87
(range, 85–90)
Sagittal Anatomic posterior proximal tibial angle Defines the relationship between the sagittal proximal tibial joint orientation line and the mid-diaphyseal line of the tibia. 81
(range, 77–84)
Distal tibia Coronal Mechanical lateral distal tibial angle Defines the relationship between the distal tibial joint orientation line and the mechanical axis of the tibia. 89
(range, 88–92)
Sagittal Anatomic anterior distal tibial angle Defines the relationship between the sagittal distal tibial joint orientation line and the mid-diaphyseal line of the tibia. 80
(range, 78–82)

Anatomic Axes
Femur: Mid-diaphyseal line.
Tibia: Mid-diaphyseal line.
Mechanical Axes
Femur: Defined by a line from the center of the femoral head to the center of the knee joint.
Tibia: Defined by a line from the center of the knee joint to the center of the ankle joint.
Lower extremity: Defined by a line from the center of the femoral head to the center of the ankle joint.
Normal values from Paley D, Tetsworth K. Mechanical axis deviation of the lower limbs: preoperative planning of multiapical frontal plane angular and bowing deformities of the femur and tibia. Clin Orthop Relat Res . 1992;280:65–71; Paley D, Tetsworth K. Mechanical axis deviation of the lower limbs: preoperative planning of uniapical angular deformities of the tibia or femur. Clin Orthop Relat Res . 1992;280:48–64.)

The center of rotation of angulation (CORA) is the point at which the axis of the proximal segment intersects the axis of the distal segment ( Fig. 26.13 ). For diaphyseal deformities, the anatomic axes are convenient to use. For juxtaarticular deformities, the axis line of the short segment is constructed using one of three methods: extension of the segment axis from the opposite side, intact bone if its anatomy is normal; comparing the joint orientation angle of the abnormal side to the opposite side if the latter is normal; or drawing a line that creates the population normal angle formed by the intersection with the joint orientation line.

Fig. 26.13, The same case shown in Fig. 26.11 with a diaphyseal nonunion with deformity. The center of rotation of angulation (CORA) and bisector are illustrated.

The bisector is a line that passes through the CORA and bisects the angle formed by the proximal and distal axes (see Fig. 26.13 ). Angular correction by rotating the bone segments around a point on the bisector results in complete deformity correction without introducing translational deformity.

Rotational deformities associated with a nonunion may be missed on physical and radiologic examination. Accurate clinical assessment of the magnitude of a rotational deformity is difficult, and plain radiographs offer little assistance. CT scanning is the best method of radiographic assessment of malrotation, as described in the next subsection.

Like angular deformities, translational deformities are characterized by magnitude and direction. The magnitude of translation is the perpendicular distance from the axis line of the proximal fragment to the axis line of the distal fragment at the level of the proximal end of the distal fragment ( Fig. 26.14 ).

Fig. 26.14, Method for measuring translational deformities. The magnitude of translation is measured at the level of the proximal end of the distal fragment.

When both angular and translational deformities are present, the CORA will be at different levels on the AP and lateral radiographs ( Fig. 26.15 ). When the deformity involves pure angulation without translation, the CORA will be at the same level on both radiographs.

Fig. 26.15, (A) When the deformity at the nonunion site involves angulation without translation, the center of rotation of angulation (CORA) will be seen at the same level on both the anterior-posterior (AP) and lateral radiographs. (B) When the deformity at the nonunion site involves both angulation and translation, the CORA will be seen at a different level on the AP and lateral radiographs.

The radiographic evaluation should also identify any compensatory joint deformities adjacent to the nonunion. These compensatory deformities are not always clinically apparent. As previously stated, failure to recognize and correct the compensatory joint deformity leads to a healed, straight bone but suboptimal functional improvement.

Radiographic analysis should therefore be performed at adjacent joints for a nonunion with deformity. In particular, for a tibial nonunion with a coronal plane angular deformity, a compensatory deformity at the subtalar joint is not only common but is commonly missed. Varus tibial deformities result in compensatory subtalar valgus deformities, and valgus tibial deformities result in compensatory subtalar varus deformities. Compensatory subtalar joint deformities are evaluated using the extended Harris view of both lower extremities, which allows measurement of the orientation of the calcaneus relative to the tibial shaft in the coronal plane ( Fig. 26.16 ).

Fig. 26.16, The extended Harris view is performed to image the orientation of the hind foot to the tibial shaft in the coronal plane. (A) The patient is positioned lying supine on the radiography table with the knee in full extension and the foot and ankle in maximal dorsiflexion. The radiography tube is aimed at the calcaneus at a 45-degree angle with a tube distance of 60 inches. (B) Anterior-posterior (AP) radiograph of the tibia shows a distal tibial nonunion with a valgus deformity. This patient had been treated with an external fixator at an outside facility. In an effort to correct the distal tibial deformity, the hind foot had been fixed in varus through the subtalar joint. (C) On clinical inspection, this is not always obvious and can be missed. (D) The extended Harris view of the normal left side compared with the abnormal right side. Note the profound subtalar varus deformity of the right lower extremity. Both the distal tibial valgus deformity and the subtalar varus deformity must be corrected in this patient.

Computed Tomographic Scanning and Tomography

Plain radiographs are sometimes insufficient to assess fracture healing. Sclerotic bone and orthopaedic hardware may obscure the fracture site, particularly in stiff nonunions or those well stabilized by hardware. CT scans and tomography are useful in such cases ( Fig. 26.17 ).

Fig. 26.17, (A) Computed tomography (CT) scan of the 88-year-old woman shown in Fig. 26.8 ; A shows that this fracture is in fact healed. (B) CT scan of the 49-year-old man shown in Fig. 26.8 ; B shows that this fracture has gone on to nonunion.

CT scans can be used to estimate the percentage of the cross-sectional area of bridging bone ( Fig. 26.18 ). Nonunions typically show bone bridging of less than 5% of the cross-sectional area at the fracture surfaces (see Fig. 26.18 ). Healed or healing fractures typically show bone bridging greater than 25% of the cross-sectional area. Serial CT scans may be used to evaluate the progression of fracture consolidation (see Fig. 26.18 ). CT scans are also useful for assessing intraarticular nonunions for articular step-off and joint incongruency.

Fig. 26.18, In addition to helping determine whether a fracture has united or has gone on to nonunion, computed tomography (CT) scans are a useful method for estimating the cross-sectional area of healing. In this case of an infected midshaft tibial nonunion, CT scanning is a useful method of estimating the progression of the cross-sectional area of healing over time. (A) Radiograph of the tibia 4 months after injury. It is difficult to definitively say whether or not this fracture is healing. (B) CT scan shows a clear gap without bony contact or bridging bone (0% cross-sectional area of healing). (C) Radiograph 6 months later after gradual compression across the nonunion site. (D) CT scan shows solid bony union (cross-sectional area of healing greater than 50% in this case).

Plain tomography helps evaluate the extent of bony union when hardware artifact compromises CT images.

Rotational deformities may be accurately quantified using CT by comparing the relative orientations of the proximal and distal segments of the involved bone to the contralateral normal bone. This technique has been mostly used for femoral malrotation but may be used for any long bone.

Nuclear Imaging

Nuclear imaging studies are useful for assessing bone vascularity at the nonunion site, the presence of a synovial pseudarthrosis, and infection.

Technetium-99m-pyrophosphate (“bone scan”) complexes reflect increased blood flow and bone metabolism and are absorbed onto hydroxyapatite crystals in areas of trauma, infection, and neoplasia. The bone scan will show increased uptake in viable nonunions because there is a good vascular supply and osteoblastic activity ( Fig. 26.19 ).

Fig. 26.19, Technetium bone scanning of three different cases showing (A) a viable nonunion, (B) a nonviable nonunion, and (C) a synovial pseudarthrosis.

A synovial pseudarthrosis (nearthrosis) is distinguished from a nonunion by the presence of a synovium-like fixed pseudocapsule surrounding a fluid-filled cavity. The medullary canals are sealed off, and motion occurs at this “false joint.” Synovial pseudarthrosis may arise in sites with hypertrophic vascular callus formation or in sites with poor callus formation and poor vascularity. The diagnosis of synovial pseudarthrosis is made when technetium-99m-pyrophosphate bone scans show a “cold cleft” at the nearthrosis between hot ends of ununited bone (see Fig. 26.19 ).

Radiolabeled (e.g., indium-111 or technetium-99m hexamethyl propyleneamine oxime [HMPAO]) polymorphonuclear neutrophils (PMNs) accumulate in areas of acute infection, so these scans are used for evaluating acute bone infection.

Gallium scans are useful for the evaluation of chronic bone infections. Gallium-67 citrate localizes to sites of chronic inflammation. The combination of gallium-67 citrate and technetium-99m-sulfa colloid bone marrow scans can clarify the diagnosis of a chronically infected nonunion.

Other Radiologic Studies

Fluoroscopy and cineradiography (see Fig. 26.4 ) may be needed to determine the relative contribution of a joint and an adjacent nonunion to the overall arc of motion. Fluoroscopy is also helpful for guided-needle aspiration of a nonunion site.

Ultrasonography is useful for assessing the status of the bony regenerate during distraction osteogenesis. Fluid-filled cysts in the regenerate can be visualized and aspirated using ultrasound technology, thus shortening the time of regenerate maturation ( Fig. 26.20 ). Ultrasonography can also confirm the presence of a fluid-filled pseudocapsule when synovial pseudarthrosis is suspected.

Fig. 26.20, (A) Radiograph of a slowly maturing proximal tibial regenerate. (B) Ultrasonography shows a fluid-filled cyst (arrow) .

Magnetic resonance imaging (MRI) may occasionally be used to evaluate soft tissues and cartilaginous and ligamentous structures.

Sinograms may be used to image the course of sinus tracts in infected nonunions.

Angiography provides anatomic detail of vessels as they course through a scarred and deformed limb. Although unnecessary in most patients presenting with a fracture nonunion, angiography is indicated if the viability of the limb is in question.

Preoperative venous Doppler studies should be used to rule out deep venous thrombosis in patients with a lower extremity nonunion, who have been confined to a wheelchair or bedridden for an extended period. Intraoperative or postoperative recognition of a venous thrombus or an embolus in a patient who has not been screened preoperatively does not make for a happy patient, family, or orthopaedic surgeon.

Laboratory Studies

Routine laboratory work, including electrolytes and a complete blood count (CBC) , are useful for screening general health. The sedimentation rate and C-reactive protein are useful in regard to the course of infection. If necessary, the nutritional status of the patient can be assessed via anergy panels, albumin levels, and transferrin levels. If wound-healing potential is in question, an albumin level (≥3.0 g/dL preferred) and a total lymphocyte count (>1500 cells/mm 3 preferred) can be obtained. For patients with a history of multiple blood transfusions, a hepatitis panel and a human immunodeficiency virus (HIV) test may also be warranted.

When infection is suspected, the nonunion site may be aspirated or biopsied under fluoroscopic guidance. The aspirated or biopsied material is sent for a cell count and Gram stain, and cultures are sent for aerobic, anaerobic, fungal, and acid-fast bacillus organisms. To encourage the highest yield possible, all antibiotics should be discontinued at least 2 weeks before aspiration.

Consultations

Many issues commonly accompany nonunion, including soft tissue problems, infection, chronic pain, depression, motor or sensory dysfunction, joint stiffness, and comorbid medical problems. A team of subspecialists is usually needed to assist in the care of the patient, beginning with the initial evaluation and continuing throughout the course of treatment.

A plastic reconstructive surgeon may be consulted preoperatively to assess the status of the soft tissues, particularly when the need for coverage is anticipated after serial débridement of an infected nonunion. Consultation with a vascular surgeon may be necessary if the viability (vascularity) of the limb is in question.

An infectious disease specialist can prescribe an antibiotic regimen preoperatively, intraoperatively, and postoperatively, particularly for the patient with a long-standing infected nonunion.

Many patients with nonunions have a dependency on oral narcotic pain medication. Referral to a pain management specialist is helpful both during the course of treatment and in ultimately detoxifying and weaning the patient off of all narcotics.

Depression is common in patients with chronic medical conditions. Patients with nonunions often have signs of clinical depression. Referral to a psychiatrist can provide great benefit.

A neurologist should evaluate patients presenting with motor or sensory dysfunction. Electromyography and nerve conduction studies can document the location and extent of neural compromise and determine the need for nerve exploration and repair.

A physical therapist should be consulted for preoperative and postoperative training with respect to postoperative activity expectations and the use of assistive or adaptive devices. The goals of immediate postoperative (inpatient) rehabilitation include independent transfers and ambulation, when possible. Outpatient postoperative physical therapy primarily addresses strength and range of motion of the surrounding joints but may also include sterile or medicated whirlpool treatments to treat or prevent minor infections (e.g., external fixation pin sites).

Occupational therapy is also useful for activities of daily living and job-related tasks, particularly those involving fine motor skills such as grooming, dressing, and use of hand tools. Occupational therapy may also provide adaptive devices for activities of daily living during treatment.

A nutritionist may be consulted for patients who are malnourished or obese. Poor dietary intake of protein (albumin) or vitamins may contribute to delayed fracture union and nonunion. A nutritionist may also counsel severely obese patients to reduce body weight. Obesity increases the technical demands of nonunion treatment and the risk of complications.

Anesthesiologists and internists should be consulted during preoperative planning for the elderly or patients with serious medical conditions to decrease the risk of intraoperative and postoperative medical complications.

Treatment

Objectives

Obviously, treatment is directed at healing the fracture. Healing, however, is not the only objective as a nonfunctional, infected, deformed limb with joint pain and stiffness will be an unsatisfactory outcome for most patients even if the bone heals solidly. Emphasis is, thus, on returning the extremity and the patient to the fullest function possible during and after treatment.

Treating a nonunion can be likened to playing a game of chess; it is difficult to predict the course until the process is under way. Some nonunions heal rapidly with a single intervention. Others require multiple surgeries. Unfortunately, the most benign-appearing nonunion occasionally mounts a terrific battle against healing. The treatment must, therefore, be planned so that each step anticipates the possibility of failure and allows for further treatment options without burning any bridges.

The patient's motivation, disability, social and legal issues, mental status, and desires should be considered before treatment begins. Are the patient's expectations realistic? Informed consent before any treatment is essential. The patient needs to understand the uncertainties of nonunion healing, time course of treatment, and number of surgeries required. No guarantees or warranties should be given to the patient.

If the patient is unable to tolerate a potentially lengthy treatment course or the uncertainties associated with the treatment and outcome, the option of amputation should be discussed. Although amputation has obvious drawbacks, it does resolve the medical problem rapidly and may, therefore, be preferred by certain patients. It is unwise to talk a patient into or out of any treatment; this is particularly true for amputation.

When feasible, eradication of infection and correction of unacceptable deformities are performed at the time of nonunion treatment. When this is not practical or possible, the treatment plan is broken into stages with the following priorities:

  • 1.

    Heal the bone.

  • 2.

    Eradicate infection.

  • 3.

    Correct deformities.

  • 4.

    Maximize joint motion and muscle strength.

These priorities do not necessarily denote the temporal sequencing of surgical procedures. For example, in an infected nonunion with a deformity, the treatment may begin with débridement to eliminate infection, but the overriding priority remains to heal the bone.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here