Open Fractures

Direct Blow

A direct blow causes a local area of injury with limited extension. The open wound can be from the direct-blow site, causing an “out-to-in” mechanism, or the potentially contra-coup injury opposite from the direct-blow site with an “in-to-out” injury ( Fig. 18.1 ). Both injuries are serious, but the direct site may be more contaminated and have a more serious associated soft tissue injury. Furthermore, the site may have an injury in evolution that may worsen before improving.

Crush Injury

Crush injuries create immediate and sometimes irreversible associated soft tissue injury. Epidermolysis usually evolves an enlarging and worsening pattern consistent with the energy of injury imparted to the limb. The unsalvageable skin affected, and many times the underlying degloved fat, needs radical débridement. If prolonged or severe enough, the injury will be similar to an internal amputation with an associated open fracture of varying severity. The fracture can be a simple to a complex pattern. The circumferential crush injury creates problematic limb-salvage options. Complete musculotendinous, venous, and cutaneous disruption can be present, requires assessment, and may worsen with time ( Fig. 18.2 ).

Explosion and Blast Injury

The pathophysiology of explosion or blast injuries is dependent on the force and location of the source, and it evolves with time. It starts with the detonation followed by the blast wave, blast wind, and the anatomic stress wave. The detonation is from a high-speed chemical decomposition of an explosive gas. Space occupied by the explosive is now occupied by gas under high pressure and temperature. The blast wave is a pressure pulse a few millimeters thick that travels at supersonic speed radially from the center of the blast. The leading edge rapidly decreases in pressure and becomes an acoustic wave. The local effect is a positive-pressure destructive shock wave followed by a negative-pressure wave. The pressure drops below ambient pressure, and a vacuum effect takes place. The mass movement of air causes a blast wind that can propel objects and people considerable distances. The anatomic stress wave caused by blast-wave interaction with a person with local overpressure has increasing pressure up to eight times normal. The stress wave causes rapid acceleration of the body surface and a stress wave. The positive-pressure shock wave creates immediate muscle damage, whereas with the negative-pressure shock wave, it takes time for the full destructive pattern to evolve and be determined.

Three blast injuries are noted: primary, secondary, and tertiary. The primary blast wave is caused by the direct effect of the blast wave on the body. The effect is dependent on the distance from the source. The lethal radius is three times greater in water. It is increased at the reflecting surface. The injury is seen almost exclusively in air-filled structures. The ear is the most sensitive, but the respiratory system is the most common cause of morbidity and mortality. The gastrointestinal tract is the most common cause of delayed morbidity and mortality. Major limb amputation occurs as a result of the blast-wave–induced fracture followed by the blast wind avulsing the fractured limb. The secondary blast injury occurs from the casualty being struck by fragments from the explosive device or by secondary missiles being energized by the blast. This injury type has the same principles of diagnosis and care as for bullets or open wounds. Flying casing fragments and debris are irregularly shaped and less aerodynamically stable. The drag slows the fragments’ speed; therefore they travel shorter distances. The fragments can tumble upon contact with tissue, thus causing potentially greater tissue damage than a bullet. A higher risk of environmental debris being dragged into the wound and causing greater contamination is present. A large, slow-moving projectile may crush a larger amount of tissue. Missile fragmentation can increase temporary cavitation effects. The tertiary blast injury occurs when the victim is thrown against the ground or solid objects. The injuries are similar to blunt trauma or falls. Care follows blunt trauma guidelines. The tertiary blast wave causes fractures, crush injuries, amputations, and associated lacerations as people tumble and impact stationary objects ( Fig. 18.3 ). Tourniquets or modified objects to create tourniquets are important to avoid exsanguination. Therefore a blast injury can present with a spectrum of acuity (blast injury with all three components in varying degrees) and injuries (thermal, chemical, biologic, and multisystem). Patients require a multidisciplinary team secondary to the obvious and subtle injury patterns. Even though this injury is mainly associated with war injuries, terrorist (e.g., Boston City Marathon bombing in 2013) or industrial explosions could generate forces similar to those in war casualties.

Gustilo and Anderson

Veliskakis proposed the initial open fracture classification based on three types and worsening severity. Gustilo and Anderson formulated and confirmed the classification in the 1960s to 1970s. Gustilo modified the classification further. The classification is based on open tibial fractures and the size of the wound. Gustilo and Anderson determined a relationship between an increasing wound size and the risk of infection or osteomyelitis. The classification does not determine outcome or treatment.

Type I fractures have a skin laceration of less than 1 cm. This is usually a puncture wound through the skin from the bone protruding out or a direct blow from out to in. Type II fractures have a laceration greater than 1 cm and less than 10 cm. Type III fractures fall into a large and varied category, including extensive skin damage with muscle involvement, high-energy injury, crush injury, segmental or highly comminuted fracture, segmental diaphyseal osseous defect, high-velocity weapon, extensive contamination of the wound, or farmyard injury. Type IIIA fractures have a laceration greater than 10 cm, but the integument can be closed or reapproximated. Type IIIA also includes any wound size with heavy contamination with or without segmental and/or comminuted fracture patterns. Type IIIB fractures have lacerations greater than 10 cm, but the wound cannot be reapproximated and requires a rotational or free-tissue transfer for closure. Skin grafting closure does not make the wound a type IIIB. Modifications of this classification can be considered if utilizing a circular frame for treatment. If the fracture is treated with an external fixator that allows for bending and/or shortening, the wound can then be closed. This is then considered a type IIIB converted to a type IIIA wound. The amount of bending and shortening is restricted before secondary consequences such as vascular kinking and congestion can result in a limb at risk. Type IIIC fractures are open wounds with an associated vascular injury requiring repair for limb salvage ( Table 18.1 ). Gustilo also classified an open fracture that presents greater than 8 hours after injury as a special type III open fracture.

The Gustilo-Anderson classification has been able to recommend antibiotic usage based on the type of fracture. The more severe wound requires broader-spectrum antibiotic coverage. Types I and II wounds with mainly gram-positive bacteria require only a cephalosporin. Type III wounds require gram-negative coverage in addition. Contaminated wounds require penicillin for clostridium and group A streptococcus coverage. Increasing wound size and classification severity were correlated with wound infection and amputation rates. Therefore the classification was subdivided later (1970s–1980s) into three types (A, B, C) of type III injuries. The risk of wound infection was found to be as follows: type IIIA, 4%; type IIIB, 53%; and type IIIC, 42%. The risk of amputation was found to be as follows: type IIIA, 0%; type IIIB, 16%; and type IIIC, 42%. Despite the correlative increasing severity, the classification has poor interobserver reliability at only 60%. Even though this problem exists, the classification has generated worldwide acceptability. It is simple and logically stratifies open fractures. Despite being originally determined to describe open tibial fracture patterns only, it has, rightly or wrongly, expanded to classify other fractures of the body. The system does recommend methods for closure (primary, delayed, free-tissue transfer) but does not recommend overall treatment methods. Treatments do change over time and can change the classification type today. For example, vacuum-assisted closure allows us to close many wounds today (type IIIA) that would have required free-tissue transfer (type IIIB) without this method. In addition, circular external fixator frames with or without proximal corticotomy facilitate fracture manipulation to close the wound and accelerate local blood flow.

A major problem with the Gustilo-Anderson classification is related to type IIIB injuries. A 10-cm open medial wound with no muscle compartment injury and a transverse mid-diaphyseal tibial fracture pattern is not the same injury as a 10-cm open medial wound with deep posterior and anterior compartment injuries and a comminuted tibial fracture pattern with a 3-cm bone defect (critical size defect). The former injury could be treated with a tibial nail, gastrocnemius muscle rotational transfer, and skin grafting, resulting in predictable healing and low complication rates. The latter injury could be treated with a nail, antibiotic cement spacer, free-tissue transfer, and skin grafting, with predictably low healing and high complication rates, requiring multiple secondary procedures and possibly ending with successful limb salvage, unpredictable muscle and leg function, or possibly a revision amputation. Therefore, even though the Gustilo-Anderson classification is the current historical standard, it is not perfect and can be improved on for determination of treatment, outcomes, and complications.

Other Open Fracture Classifications

Tscherne and colleagues developed an open and closed fracture classification. The open types are the following: Type I is a puncture hole, in which the bone lacerates the skin from the inside out. The skin and muscle have minimal contusion. These injuries are the result of indirect injuries. Type II, moderate contamination, has circumferential skin or muscle contusion. Any fracture pattern can be present. Type III, heavy contamination, has extensive soft tissue damage to the skin and muscle and results in soft tissue problems. Nerve and/or arterial injury is common. Fractures usually have extensive comminution and devitalized fragments. He grouped all farming-related wounds, high-velocity gunshot wounds, and associated compartmental syndromes in this category. Type IV, incomplete or complete amputation, are all fractures with arterial injuries requiring revascularization. This was combined with a closed fracture, soft tissue injury classification. The types are as follows: type 0, minimal soft tissue damage with indirect violence, simple fracture pattern (e.g., torsion fracture of the tibia in skiers); type I, superficial abrasion or contusion caused by pressure from within, mild to moderate-severe fracture pattern (e.g., ankle pronation fracture-dislocation with soft tissue over medial malleolus); type II, deep contaminated abrasion associated with localized skin or muscle contusion, impending compartmental syndrome (e.g., segmental “bumper” tibial fracture); and type III, extensive skin contusion or crush, underlying muscle damage may be severe, subcutaneous avulsion/degloving, associated major vascular injury, severe of comminuted fracture pattern. The injury systems are complete, but the categories are much too variable or involve subjective discrimination. The intraobserver and interobserver agreement for the Tscherne classification are 85% and 65%, respectively.

The Arbeitsgemeinschaft für Osteosynthesefragen (AO) classification system is a modification of the Tscherne classification and employs a grading system based on the skin (I), muscles and tendons (MT), and neurovascular (NV). Each grade is further divided into five degrees of severity. This is the first system to grade the wound more on severity than just the size of the wound. In addition, it indirectly attempts to measure the amount of function based on the soft tissue injury to the muscle and nerves. The problem with this classification is the complexity of the multiple choices for different categories and therefore the inability to deploy or utilize it for daily practice or consumption.

OTA/AO Open Fracture Classification

Despite the widespread use of the Gustilo-Anderson classification, a better way to quantify the severity of open fractures is needed. For example, a Gustilo open IIIB tibial fracture with a small-sized anterior pretibial defect, no bone loss, minimal contamination, but requiring a soleus muscle rotational flap is typed the same as a Gustilo open IIIB tibial fracture with extensive degloving and contamination, greater than 4 cm segmental bone loss, loss of the entire anterior compartment, and requiring bone transport or massive autografting, free-tissue transfer, and extensive split-thickness skin grafts. To address these limitations and better define these injuries, the Orthopaedic Trauma Association (OTA), in collaboration with the AO group, created the OTA Open Fracture Study Group. With the use of the three electronic databases (PubMed, EMBASE, and Web of Science), factors utilized to evaluate open fractures of the upper extremity, pelvis, and lower extremity were compiled. Based on their clinical experience and the existing literature, seven fellow-trained orthopaedic trauma surgeons independently examined and prioritized factors for inclusion or exclusion for this new open fracture classification ( Table 18.2 ). A rank-order mean for each factor was calculated and measured as to its relative importance ( Table 18.3 ). The other factors were simplicity, pathoanatomy, the exclusion of systemic issues, and the anatomic characteristics of the injury. This group recently finalized a new open fracture classification system to facilitate consistent application and communication in assessment, treatment, and research. The new OTA/AO Open Fracture Classification (OFC) includes five assessment categories: (1) skin defect, (2) muscle injury, (3) arterial injury, (4) bone loss, and (5) contamination, with each category subdivided into three descriptors (mild, moderate, and severe) of increasing severity ( Table 18.4 ). The classification was successfully tested for feasibility and ease of clinical data collection. The advantage of this classification is its ability to better classify the injury severity, which is a continuous variable, into different groupings of severity instead of just three categories.

Furthermore, the OFC can also be used to determine treatment implications instead of just infection as in the Gustilo-Anderson classification. A recent study prospectively evaluated 356 patients with open fractures of different areas of the body instead of just open tibial diaphyseal fractures. The use of vacuum-assisted closure (VAC), multiple débridements, antibiotic bead placement, and early amputation was evaluated. The OFC was related to the type of treatment utilized to treat this cohort of open fractures. Skin injury was the strongest predictor of VAC utilization. Skin injury and muscle injury were predictive of multiple débridements. Bone loss was a strong predictor of antibiotic bead placement. The combination of skin injury, contamination, and arterial injury was the strongest predictor of limb amputation. Further analysis will determine how these variations in the five subgroups will determine how an open fracture is treated.

At the 2013 annual OTA meeting, two research projects concerning the OFC were presented. Both have important and different take-home messages. The OFC was validated in a population of severe, limb-threatening tibial fractures. The study utilized the original prospective data collection from the LEAP study. LEAP data included the Gustilo-Anderson Open Fracture Classification, the Tscherne Closed and Open Fracture Classification, the AO Classification of Soft Tissue Injury of the Tibia (Closed and Open Lesions), the Hannover Fracture Scale, the Limb Salvage Index, the Predictive Salvage Index, and the Mangled Extremity Severity Score (MESS). The cohort data retrospectively classified the fractures using the OFC scheme. From available LEAP study data, the authors identified the most appropriate classifications and response categories from the LEAP study that would correspond to each of the five arrays of the OFC. A crosswalk between the classifications used in LEAP and the OFC was performed. As expected, each of the five OFC components showed a statistically dependent relationship with the Gustilo type and also revealed variation in the OFC classification within the Gustilo type. The polychoric correlation between each of the OFC components was low to moderate. Examining the predictive capacity of the OFC against two important outcomes assessed criterion validity: early amputation and function at 2 years posttrauma. Increased severity of each OFC component score was significantly associated with amputation. The predictive power of each OFC component with respect to amputation, as measured by predictive area under the curve (AUC), was comparable to that of the Gustilo-Anderson classification. The new OFC provides a system to classify soft tissue injuries within five clinically meaningful domains, each of which is strongly predictive of amputation, a major clinical outcome. Among salvage patients, having the highest level of the muscle, bone loss, and arterial OFC component was associated with a 2.9-, 3.8-, and 5.8-point increase, respectively, in disability at 2 years based on functional outcomes as defined by the physical and psychosocial domains of the Sickness Impact Profile (SIP). A combination criterion of the highest levels of the arterial and bone loss components was developed, which occurred in 23% of this severely injured population. The data suggest several OFC components are predictive of clinically and statistically significant differences in long-term functional outcome. Overall, the results of this analysis show that the OFC has strong content, construct, and both concurrent and predictive criterion validity. The new classification has demonstrated outstanding interobserver reliability of 93%, with improvement over the Gustilo-Anderson classification.

The other OTA presentation utilized the same LEAP data but evaluated how soft tissue injury or loss predicts amputations in severe open tibial fractures. A logistic regression model of the 19 tibial compartment soft tissue items (muscle, vein, artery, and nerve) was developed to examine their independent contributions to the risk of amputation. When included in a single model, all 19 items were able to predict amputation with an AUC of 0.833 (roughly equivalent to 80% sensitivity and specificity). Two components, the posterior tibial artery and the tibial nerve, were so highly correlated that it would have been impossible to include them as separate items in any model, and they were merged for their analysis. Using logistic regression, a subset of six items (flexor hallucis longus, peroneal artery/vein, posterior tibial artery or vein, superficial peroneal nerve, and gastrocnemius) accounted for 98% of the predictive power of the larger model (AUC = 0.815). Injury to the flexor hallucis longus muscle or any anterior compartment muscles severe enough to render them nonfunctional and a nonfunctional tibial nerve was predictive of patients who ultimately underwent amputation. As a general finding, as the numbers of individual muscles injured increased, the patient was more likely to undergo an amputation. These results may allow surgeons to more accurately counsel their patients on what to expect and enable them to maximize the predicted functional outcome for a patient with a mangled lower extremity. Therefore the OFC may be able to predict outcomes based on the severity of injury.

The OFC is a new and validated open fracture classification that allows for improved subgrouping over the Gustilo-Anderson classification. It may be able to predict operative treatment, outcomes, and potential for amputation. It will be used in parallel with, if not as a replacement of, the Gustilo-Anderson classification for future research and publications ( Figs. 18.4 and 18.5 ).

Initial Trauma Assessment

Patients with open fractures require a full trauma assessment. One should be familiar with the Advanced Trauma Life Support (ATLS) protocols. A parallel and proactive orthopaedic assessment should be performed. The mechanism required to produce an open fracture is more than the usual twisting or fall mechanism. Therefore a check for subtle, progressing, and/or serious, life-threatening injuries should be performed on all patients. An orthopaedist experienced in associated injuries, acting as a proactive and helpful consultant, is beneficial to the trauma team. Large-bore intravenous (IV) lines are required for resuscitation and antibiotic delivery. Tetanus prophylaxis is initiated as soon as possible.

Prompt Diagnosis

In the assessment of any fracture, the limb must be assessed circumferentially, with complete removal of all clothing and splints. Do not miss an open fracture. Open fractures are surgical emergencies. Determine the time of injury and facilitate all phases of patient assessment and injury imaging. A wound oozing venous blood, especially with fat droplets, is suspicious for an open fracture. Additionally, a fracture with a dried scab over the apex of the fracture could be an open fracture. Despite varying sizes of open wounds, check to make sure no skin, clothing, or foreign body is entrapped within the wound. Open wounds can have local or distant open communicating sites. Once the diagnosis is made, cover the wound with saline-soaked gauze or toppers. The saline will diminish the desiccation of the tissues. Because antibiotic, chlorhexidine, and povidone-iodine addition can affect the tissue viability, saline is preferred to other methods of gauze preparation. Once the sterile dressing is applied, do not remove it for additional provider visualization because this can be performed in the operative suite. In addition, reapplying dressings in the emergency department increases the risk of infection threefold to fourfold.

Imaging studies should be evaluated for any soft tissue abnormality or air consistent with an open fracture. Furthermore, imaging the entire bone along with associated joints will lessen the chance of missed fractures. Once the diagnosis of an open fracture is completed, appropriate treatment options and timing can be initiated.

Control Bleeding

After an initial visual assessment, apply sterile gauze or toppers to the open wound. After this, apply a mildly compressive splint to control bleeding with compression and realignment of the limb. This will also reduce pain with immobilization of the fracture and swelling with the utilization of hydrostatic forces. Remove any skin entrapped within the wound to avoid necrosis of skin edges. On rare occasions, hemostat or ligation of a bleeding vessel is required to minimize blood loss and even exsanguination. For uncontrollable hemorrhage or amputation, a temporary tourniquet can be applied and inflated proximal to the bleeding and wound ( Fig. 18.6 ). The positive effects of tourniquet utilization need to be compared with the negative effects of worsening ischemia, muscle damage, and pain.

Injury Assessment